1 1 Memo on the 2012 Gains Made by Jp Morgan Thanks to the « London

Memo on the 2012 gains made by Jp Morgan thanks to the « London Whale » event

Introduction

It will be one thing to show that the bank did make huge gains right when the “London Whale” event surged. Still it could have been a coincidence. It is another thing to prove that those gains were done thanks to this “London Whale-Loss at CIO” event specifically. The suspicion then would be easy if that was a crime novel: “see who profits from the crime” right? Beware of the smokes and mirrors now… It would be again one success to show what the close connection was between the CIO loss and the JMorgan massive gains. It would be quite another achievement to show that the bank top executives did act so that the loss at CIO would be as big as possible on well targeted positions. Even then, once this fact is ascertained, this would not prove that the bank senior management alone deliberately wanted to crash the CIO so scandalously. It would even less suggest that the bank was the one that misled investors and the markets on the way…. If the bank executives actually did so then and “in hindsight” too, they certainly bear a huge responsibility…. They might not be alone in that case though. Yet even if that had happened, it may still be possible that this course of events had indeed run out of their control and they had just tried their very best to save the bank here. And one must wonder how the regulators would have missed “that” too at the time given the intense scrutiny that they would profess with regards to their subsequent investigations. The bank executives may have been heroes actually for the country or the markets or both, acting as per the “superior interest of the nation” that was being dictated to them…. The purpose of this document is meant to address those points as granularly as possible. In particular the emphasis will be placed on the fact that this loss at CIO was under full control and profitable for the bank right through the events themselves. Some things indeed went out of control around Iksil’s communications and Drew’s reactions among other things. Not the least, the bank created as many distractions as possible for the public while observers tried to understand how profitable or un-profitable the whole operation had been. Carl Levin loudly complained that the light had not been made at all on the matter in September 2013. This conclusion of Levin prevails in September 2017 actually. Is it too late?

This document may sound very “technical” from the start. It is needed unfortunately so in order to reconcile the facts with the accounting ledgers of the bank and dissolve the crowd of decoys that had been planted here on the front stage all along. For the sake of completeness the reader will first find below the many topics that have been analyzed. Yet, on what follows, not all of those items will be developed as otherwise the document would have taken really hundreds of pages. So there will be this quick “summarized plan” for the experts to see the overall framework that led to this memorandum on the gains that JpMorgan did make in the course of the “London Whale” event. Next there will be a detailed description of what happened and how this combined with key inferences to lead to a salient conclusion. The reader next should usefully read the “Var History” document. Thus a full picture will be accessible as to the last key point that is to determine whether the bank top executives did construct all this manipulation on purpose, aside from the bits and pieces that went out of their control in the process as they say.

Summarized Plan: the essentials (they will not be all depicted here)

1- How to read through the 10-Q and 10-K reports? a. ‘identical underlying’, Gross amounts, netted amounts, ‘fair value’, ‘carrying value’ i. What is audited (net interest income) and what is not audited (trading related income) 1. Main reference tables present in the 10-Q reports a. GCB, DCM, CIO, treasury, liquidity reserve (not developed much) i. Mark to market, Measuring Fair value and fair value election

2- What is the actual bonanza brought up by the “London whale” event? a. Clearing the legend: Table on gross P&L balances i. Tangible equity history: the Ariane Thread since 1999 1. SFAS107 history and “other collateral”: the root cause for the scandal a. PWC vs JPM: the ongoing mismatch that cleared in 2012 i. Cost of assets and liabilities that eased in 2012 1. “other assets” that shrank in 2012 3- How did the loss spread into the book in 2012? In a well organized way actually a. Effect of Dec 15th 2011, next Feb 9th, March 12th, April 6th, May 10th 2012 i. Balance of losses during H1 2012: the trades were scrutinized 1. Drift of IG9 forward spread in 2012: the trades were well executed overall a. Long term history of skew: the trades were sensible i. Simulated Recovery: the trades were NOT flawed 4- How did the bank report the event itself? a. The actual $6 billion loss attribution: 100% senior management subjectivity i. RFS: complete distraction 1. Deferred tax “benefit” trick: what a unique reporting error a. Maiden Lane: playing with calendars i. Provisions: increased not released 5- Conclusion: the bank made huge gains, acted on purpose all along and hid them as much as possible a. The year 2012 vs the others….A damn good year! i. Initial descriptions of the firm: misleading at best 1. Restated numbers (through the restatement): what about 2011 then? a. Final comment on descriptions and restatement; that was a “tempest in a teapot” when all is told…

Developed plan now……

1- How to read through the 10-Q and 10-K reports? a. ‘identical underlying’, Gross amounts, netted amounts, ‘fair value’, ‘carrying value’

The reference document here is the 10-Q report filed on may 10th 2012 for the first quarter of 2012

The very start of this document is tough but necessary to grasp the picture of the backbone of the “London Whale” scandal. These are just 4 lines right below here that one should really take the time to understand…..This will open the way to address the question: “how did the performance of the “tranche book” of CIO (the SCP apparently for the bank) enter the firm-wide valuation process?”. 4 Stages will be shortly pictured on the follow: the “gross notional” step, the “netting” step, the “risk modeling based on identical underlying” step, the ALCO or “senior management” step. These four lines are describing the root of all the scandal that will surface under the banner “the London Whale”. The content will certainly look quite abstract and the reader is invited to revert as often as necessary in order to see that those 4 lines describe the “skew risk”.

The “skew risk” is what bothered Dimon and the regulators as early as 2005. It had paved the way of JpMorgan the legendary Investment Bank for a decade already. It is the risk that is inherent to the CDS markets from their very start back in 1994. It is what will induce the creation of CIO in 2005. It is what will spark the birth of the future “tranche book” within CIO in 2006.Yet this big book could NOT have a name. This was also due to the very nature of the “skew risk” that is actually described opaquely in these 4 lines below (legal documentation issue). This “skew risk” is ALSO what will be the common denominator behind just all the hedging strategies

3 that will be deployed through the “tranche book” of CIO since 2007. It is also the “skew risk” that will spark the financial crisis in 2008 predictably so. It is what will prompt the “exotic credit wind down” plan of Dimon in 2010. It is what will delay for few more months in 2012 the quite seamless transfer of the “tranche book” towards the IB as a preliminary step to finalize this “wind down” plan. This delay here and the skew were known as such by all the regulators in late 2011. It is the price of this “skew risk” through the IG9 10yr index in 2012 that will determine the instant reported “tangible capital gain” that the firm will record starting in August 2012. The “skew risk” while being so truly central is nevertheless the risk that the bank and all the regulators will keep away from the public sight all along those years. One has just these basic 4 lines below…..

There is just one Keyword in the “London Whale” scandal that matters : “Identical underlying” :

”(c) Represents the total notional amount of protection purchased where the underlying reference instrument is identical to the reference instrument on protection sold; the notional amount of protection purchased for each individual identical underlying reference instrument may be greater or lower than the notional amount of protection sold. “

The table above shows that about $3 trillion of “credit derivative” (or CDS for the sake of simplicity) protection is purchased on one type say “A” of CDS and an almost equivalent amount of $3 trillion of CDS protection is sold on a different CDS type called say “B”. These $3 trillion exposures are to sit one in front of the other every day at JpMorgan. They are NOT to be “fungible” in principle, ie to be collapsed altogether in the form of a much reduced net exposure any time soon in real life. Yes they both type “A” and type “B” refer to what is called the same “reference obligation” or “identical underlying”…. And yet they are NOT always legally fungible. For example they are NOT all mechanically “enforceable” in “netting agreements”. For example too indices and their related tranches are separate. Indices and their single name constituents are also separate. The very same CDS contracts may as well be either held in an ISDA Master Agreement, or be held in a “separate ISDA agreement”, or in specific “client servicing” contract with the bank, or also in a standalone synthetic securitization vehicle (SPEs or VIEs in jargon) or else…. As to the so called ‘bespoke’ transactions, which can be considered at times as ‘credit related notes’, the same principle applies. The ‘bespoke’ tranches are separate from index tranches, indices, or the single name CDS from a pure legal standpoint as they involved the firm “credit” in a very peculiar way.

All this may sound like another alphabet soup. It is more than that as every contract gives way to diverse claims that may not be reconciled so smoothly in crisis times. The case of the Russian CDS on GKOs and the demise of

LTCM are strong reminders of that since 1998. The contractual “basis risk” here does lead to major financial and economic crisis as the year 2008 showed. Are we really out of the woods in that regard in 2017? See the central bank rates, see the public debt growing size and make your conclusions… Yet all these different forms of CDS contract may refer to the very same “identical underlying” as JpMorgan puts it, and the problem remains. In day-to-day practice it means that market counterparties do NOT aggregate all those categories of CDS holding say under a series of one single name CDS and thus into one price based on a common ‘reference instrument’. This is not “legally enforceable” and therefore this is NOT what the regulators would accept in light of 1998 and 2008 to say the least. But the risk modeling systems of Jp Morgan do make the aggregation at one stage if only to process a comprehensive attribution of performance through the valuation process of the bank across the diverse business units involved using all kinds of contracts. Why is that? Well simply because the bank has to measure its aggregate risks per “identical underlying” exposure. This is just good common sense risk management 1-0-1 and ALSO a requirement of regulators to do so anyway.

But each contract has a specific price despite the common “identical underlying” exposure, knowingly so. Thus the bank has to account for the “basis risk” that translates into slightly different prices all the time. The market counterparties to the bank Jp Morgan, and Jp Morgan itself, thus have separate pricing sources at the start and, although the positions can be compensated in-house through risk models based on their common reference to the same “reference obligation”, the diverse CDS usually are NOT “designated as hedges”. Yet they all have to be reported at their ‘fair value’ under the ‘mark to market protocol’ and in compliance with the firm policy of valuation in force. This implies different “consensus prices” for each type of contract involved. This mandates a stringent reconciliation process if only to avoid a Ponzi Scheme build-up whereby lasting price differences on the very same contract generate a fictitious gain inside the same firm. This ALSO implies on the follow a “netting” process internally so that the bank can assess what its net exposure is per contract and next per “identical underlying” risk. Please notice that the $3 trillion figure is shrunk down to $129 billion here after netting….The reduction in size of such magnitudes (10 times over $3 trillion) would bring up a significant but undue economic result anytime one price difference is missed.

What happens here in practice through the netting process? What is the “value” involved in the structural price differences inherited from the legal contractual “basis risk”? Let’s make here a quick simplistic numerical example to grasp how much the “basis risk” or any undue price difference matter in the accuracy of a netting process of this magnitude. It will be based on these $3 trillion of notional amount that once netted get reduced to $100 billion or less….Let’s imagine that the firm JpMorgan has say $3 trillion of one standalone “basis risk” potentially. In practice the bank sold $3 trillion of protection on “contract A” and purchased $2.9 trillion of protection based on the “identical underlying” through a “contract B” this time. “A” and “B” are two contracts that have the very same quotes since they have the very same wording. BUT they are not stored in legally enforceable netting agreements. Thus the bank gets 2 sources of prices for “A” and “B” independently. And the counterparties are different: one for “A” and one for “B”. The bank has to reconcile internally the price for “A” and the price for “B” as the example here will show. It matters critically for Jp Morgan at least to the very finest details so that the $3 trillion notional amount is shaved down to $100 billion or less.

The bank after netting is left thus with a net $100 billion of outright exposure selling protection for the sake of this example. Let’s assume that the average spread is 100 Bp and that the “tenor” or ‘maturity left’ is 5 years. These are typical features of what the bank exposure is in general. It is made simple again here for the sake of clarity. Let’s now imagine that there is a 1bp of basis risk between “A” and “B” simply because these contracts are not held in a common netting agreement that is legally enforceable. In simple terms, the bank has to make its assessment. A 1 BP has showed up by accident. Fortunately in the example there is just one price difference to assess. What would happen if this tiny 1Bp is not properly assessed? Let’s assume that this 1BP of “basis risk” here is thus NOT reconciled properly through the netting process between the price of the protection sold and

5 the price of the protection purchased elsewhere in the bank. This unchecked 1 Bp is say in favor of the bank here. A gross calculation based on the 1BP reportedly “gained” on $3 trillion leads to a positive revenue for the bank of “$3 trillion x 5 x 0.01%” or $1.5 billion. The bank produced a $5 net profit per quarter. Thus this undue gain weighs 30% of the reported quarterly profits. How much would the bank expect to make on the net $100 billion at 100 bps if the position say was already at a very significant gain of 20 bp in mark to market? As such this would be quite an achievement in trading terms. Another gross calculation would yield “$100 billion x 5 x 0.2%” or $1 billion. Thus this “nasty” 1bp of unchecked “basis risk” here obscures any major outstanding gain or loss directly resulting from massive market moves impacting trading performance for real… This is totally unacceptable. This example just showed how critical the stringent reconciliation of prices is INSIDE JpMorgan when netting the derivatives exposures considering that the magnitudes at play are this big at $3 trillion and $100 billion ‘net’.

One may argue that the bank has many hundreds of different basis risks at the start. Granted… Thus each basis risk alone as such would not be a big impact on earnings. But it remains that they altogether convey a “ bp” basis risk that is worth $1.5 billion at Jp Morgan. Thus the many hundreds of basis risks only require Jp morgan to assess stringently all of them as the order of magnitude for the ultimate inaccuracy on the earnings remains the same, ie about $1.5 billion. And the bank JpMorgan reports about $5 billion earnings per quarter then.

This basis risk “value” must therefore be assessed AND scrutinized. Otherwise the earnings at Jp Morgan are going to be pretty inaccurate from the very start of the valuation process. This must be done NOT ONLY by the bank BUT ALSO by regulators and the clearing entities like ICE day after day since the early days of 2012. So at some stage indeed, the bank reconciles the different prices for the different contracts that refer to the same “identical underlying” risk. And next the bank must scrutinize every corresponding basis risk. It matters a lot to notice how much the $3 trillion notional figure is reduced because the performance of the bank is critically measured and impacted from the change of the net figure quarter after quarter based on the ultimate ‘fair value’ of those gross notional amounts (see the example above). To reach this ultimate “fair value” for each type of contract, the bank has started with as many notional amounts, with as many different prices and had shrunk all this into one standalone net exposure per “identical underlying” risk using a stringent reconciliation process to eliminate any undue price difference and to value at best the basis risk itself. How does the bank “value” this basis risk in order to properly “price” the resulting “identical underlying” risk itself? All this stage is based again upon a proper assessment of the inherent “price uncertainty” prevailing at the time.

The bank provides 2 other key footnotes on the matter set above with reference to the table displayed here. First the footnote “(d)” states that “(d) Does not take into account the fair value of the reference obligation at the time of settlement, which would generally reduce the amount the seller of protection pays to the buyer of protection in determining settlement value.” This means that the notional amounts are NOT computed based upon the price of the “common underlying” paradoxically enough. The “common underlying” is usually priced based on a “reference obligation”. The short explanation for this choice described in the footnote (d) here usually is that the price and liquidity of the “reference obligation” are much worse than the ones of the single name CDS itself. This can only mean that the aggregation model at Jp Morgan is inferring this notional amount based on the CDS contract of either type “A” or type “B” or both actually while these CDS prices differ in general by their different nature and therefore in quantity of intrinsic liquidity. This means that there is a “basis” risk that exists anyway between the price of the ‘reference obligation’ and the price of the CDS for the same ‘reference obligation’. This difference in price does affect the associated outstanding net exposure. There is no clarity as to which price among the diverse CDS contract prevails. But the price itself of the “reference obligation” is NOT applied for sure.

Thus the net notional amount and the reported performance depend also on the price differences existing between the CDS contracts themselves. This means that the ‘basis’ risks between the ‘cash underlying asset” and the ‘CDS’ is NOT used and therefore is not fully “cleared”. What is used is the compounded effect that

6 shows from one CDS contract to another CDS contract and is often called a “skew” risk existing between a CDS of type “A” and a CDS of type “B”. As such the “value” of the netted notional amount for the “identical underlying” will still be just a proxy since it will be based solely on derivatives and not their common “reference obligation”. Based upon the former example above one has the magnitude of the impact of the “basis” and the “skew” on earnings if they are NOT finely tuned quarter after quarter. Therefore a “basis risk” will still be left pending between the “reference obligation” and the crowd of diverse CDS contracts gravitating around it all along. And senior management shall have to make a decision on the matter every quarter at least.

The footnote “(e)” now provides the main different sources of basis and skew risks: single name CDS (type ”A”), “credit indices” (say type “B”), and “credit related notes” (say type “C” which likely includes all the tranches based either on single name CDS or credit indices). The latter type “C” is actually a huge conveyor of “skew risks” as it includes the ‘bespoke’ transactions. “(e) Represents protection purchased by the Firm through single-name and index CDS or credit-related notes.” It remains that some contracts are “legally enforceable” and others are not that conveniently netted…

The gross “receivables” and “payables” that are reported BEFORE netting through “enforceable” contracts are disclosed in the 10-Q reports

A few pages before the first table displayed here (go to page 104 in this10-Q of May 10th 2012) one can see another synthetic table confirming the collapse of the skew and basis risk and the absence of designation as “hedge” for CDS in general (see the ‘credit’ derivative line) at jp Morgan:

One could see that from the former table in the blue circled figures one had to net -$206 614 million and $61 849 million to get to the net “protection” sold as per Jp Morgan risk models which were already incorporating the basis risks in an undisclosed fashion. How much money was involved here in the “basis” risk? Who knows? The net amount was $144 765 as a potential “gross receivable” notional amount. But on the table right above here as circled in, the net amount of “receivables” is only of $126 555 million. The difference is likely due to trades which are deep in loss and still alive for whatever reason. Although they should provide “receivables” the current mark to market price for these positions brings up a loss that is larger than the total one expected from the receivables in question. At least this is one sensible assumption to make for want of choice. That is worth of $18 billion… Not a small fry as pointed out in the first example above…When one sells protection on CDS indeed, one expects “receivables” on those derivatives positions in the form of premiums similar to what an insurer would get from his clients. Here in the table right above, one can see that Jp Morgan has “only” $126 555 million. This is close to an $18 billion reduction from the gross “netted” receivable amount of $144 765 million. This is most likely due to some price changes that occurred between the inception of some trades and the current applied “mark to market” “mid price”. This is a manifestation of real market price changes when applied to many open trades. And the question remains: “what about the value of the basis risk here?”...

One could not tell indeed whether this is a simple price change occurring on a standard single trade or whether this is a just “basis” risk effect, ie small price changes occurring between CDS categories which refer to the same “underlying instrument” somewhat but differ legally speaking. It could be both actually. Who knows?

But what is even more interesting here is that the ultimate “net derivative receivables” for credit is only of $6 625 million. Pfew! From $3 trillion down to $7 billion…. The sensitivity of JpMorgan to any price difference is just bigger than described before. Needless to make another example based on those magnitudes here to emphasize further the need to make quite stringent reconciliations across the many internal prices sources to estimate at best next the “basis risks” and “skew risks” on credit derivatives. One can notice at the bottom of the table that this $6 625 million amount enters as part of the “fair value” among “trading assets and liabilities”. The number goes straight into the books and records of the firm…. One will notice that the ‘net derivative payable’ amount for credit derivatives is $6 703 million. If one added the net receivables and the net payables, the “net fair value” for credit derivatives seemed to be then $6 625-$6 703=-$78 million. Hmmm…Same comment as few lines above….

Thus one has gone down from a $3 trillion figure to a $78 million net figure where likely an $18 billion ‘mark to market’ value for some trades is at stake…..Therefore one may summarize saying that the bank shrinks $3 trillion down to $78 million net on credit derivatives by means of “netting” all the diverse contracts using many different price sources for the very same contract and carrying many “basis risks” from one contract to the next…At the very least every price difference has to be scrutinized, reconciled and adjusted. The 1Bp for an average credit spread of 100 Bp is worth $1.5 billion at JpMorgan. Is that equivalent to 30% of the quarterly earnings of the big US Bank? Is there any liquidity reserve for the uncertainty carried through all these prices? There is probably not such a big reserve to take as in fact this uncertainty is NOT supposed to be left unchecked normally. It does NOT have to exist IF the reconciliations are done normally.

Yet although the “basis risks” is just everywhere, it is undisclosed. This negative $78 million net figure is really to be contrasted first with the ‘basis and skew’ effect mixed with the market price changes that one can grossly proxy as being worth something like $18 billion here (difference between $144 765 million and $126 555 million). The net -$78 million itself conveys the same mix of “basis risk” but only with the “legally enforceable” netting. What about those other contracts that are NOT “legally enforceable”?

It matters now to remember that this -$78 million will be restated in July 2012 by some $660 million due allegedly to unknown internal “price differences” on credit indices and related tranches. One wonders how they could ever have been missed. But although the bank shall state that this was a pure price difference spotted at the CIO London, one will not make the light of the influence of the “basis” risk in general in all this. It remains undisclosed in full…One sure thing is that the net “fair value” is directly impacted either by unchecked price differences on the very same contract or by those many “basis risks” as the netting process based on “identical underlying” is run. The “net-able” contracts or in-house procedures must address any price difference in the first place for aggregation. Any “fair value” change as a result impacts the performance of the firm through the “books and records”. Thus any price difference does impact the “fair value” determination in the netting process. If one shrinks $3 trillion down to $78 million at Jp Morgan, any “basis risk” requires maximum attention, any original price difference mandates the very same stringent scrutiny.

Given the orders of magnitude at play, it matters to look deeper into how the different CDS contracts are netted. The firm provides a cryptic but useful footnote:” (c) As permitted under U.S. GAAP, the Firm has elected to net derivative receivables and derivative payables and the related cash collateral received and paid when a legally enforceable master netting agreement exists.” This sentence provides a straight reference to the actual valuation policy of the firm itself, under the US GAAP standards. The footnote points to the surprising “Net

8 amount” of about $6 625 million that is so much lower than the $126 555 million amount. This is a massive reduction of money “owed” or “to be received”. This is due to the presence of “collateral” and “legally enforceable master netting agreement”.

This valuation process involves the view of the many market counterparties facing Jp Morgan on its credit derivatives trades. The bank first netted the similar contracts altogether, obtained a net value and next netted this first net value with the value of the collateral holdings. As the original example tried to show, the order of magnitude involved mandates a very accurate process. It is meant to apply only one price per CDS contract across the whole firm, irrespective of whether the different business units selected initially different prices on their side. More the price of the collateral holdings are to be checked as well stringently. Counterparties post and receive collateral daily with JpMorgan in the critical context of the margin calls. This clearly indicates that market counterparties ALSO have their say here on every “basis risk” or on every other price difference. This sentence thus unveils a critical stage of the mark to market that has always been based upon a “consensual” valuation.

The counterparties do have their say in the internal reconciliation process. If JpMorgan ever was complacent towards its price differences existing between CIO and the IB, its counterparties were NOT. Here there is just NO room for “complacency”, “deficient control”, “lasting price differences”, ignored “basis risks”….. Real money is at stake day after day. As part of the consensus building, the market players pledge some assets to their counterparty to compensate for some money they owe their market counterparties on their trades while they want to keep the derivative trade alive. What is their starting point for all? A “consensus mid price” for all….Of course the value of those collateral assets are re-valued as often as the derivatives positions are, ie daily for index and tranche CDS in 2012. Almost all the CIO positions and trades of the “tranche book” are in this case. This netting occurs actually right after a mark from the market price has been set to every positions. This is when one price for all business units is applied for the netting per contract to proceed across the whole firm according not only to the internal “view” of the firm. But it mostly is finalized according to the claims of the market counterparties of the firm. As such the latter claims define a consensus that usually is challenged by other consensuses like the one built by ICE or/and the one built by MarkIT and Totem. This is when the price differences are reconciled and flattened out through the means of many adjustments. The front office staff usually never sees these adjustments in mark to market. Here the process operates as prescribed by the SEC (1992), the OCC (1993), Jp Morgan and Paul Volcker and the “group of 30” (1993).

This stage of netting thus happens before the risk models of Jp Morgan revalue entirely every single credit derivatives based upon its group of “identical underlying” with the view to re-allocate performance against other assets and liabilities like loans, debt, pension fund, health care contribution, deferred compensations, equity, forex, commodity, real estate, private equity, and so on, and so on… Here the firm MUST make its own confidential assessment as to what the average “basis risk” is between CDS contracts and the “reference obligation”. The firm ALSO then determines what its view is on the “skew risks” spreading across the different types of CDS contracts traded by the business units in the firm. And if one wonders whether actually the firm actually does it, one must consider the cases of bankruptcy themselves involving “reference obligation” and different CDS contracts related to it altogether. When an event like this occurs, ie a bankruptcy, or when a company restructures its debt, the price of the “reference obligation” is not zero but it is very far from 100% of its original value. The different CDS contracts have other prices that may differ a lot. This induces big price differences in the liquidating value of the “reference obligation”. Indeed a lot of legal scenarios become possible in a context where a lot of uncertainties still prevail. And the ultimate liquidation price depends on the contract itself at the end of the day, with all associated costs being included.

The bank has the legal obligation to know its contractual exposure then for sure. Lawyers are involved within the firm. One can imagine easily that the bank ALSO must try to pre-empt any surprise looking forward with all its other exposures based on the diverse “reference obligations” that sit on the bank’s balance sheet directly or indirectly. In the course of 2008 and 2009, many bankruptcies occurred, the Jp Morgan lawyers were involved, and they would set the rules between CIO and the IB who had different ways to project their respective exposures to events of bankruptcy or restructuring. They were directly involved in the assessment of every basis risk standing between CIO and the IB. Someone like Jeremy Barnum (Jp Morgan MD at the IB at the time) could well testify on that. In late 2010 the “restructuring” of Thomson’s Debt proved that the basis risk and

9 therefore the skew risk were dangerous even for a standalone event as it may induce an unwanted exposure due to this internal basis risk at Jp Morgan. Therefore the bank was scrutinizing this “basis risk” since 2009 through the legal team at least of Jp Morgan with a clear awareness of the immediate and massive economic impact that the basis risk conveyed. So it is clear that the bank had to make quite a precise assessment of this “basis risk” daily not only to set the net total exposures but ALSO to set the ‘fair value’ of these credit derivatives altogether beyond the crude market quotes or consensus mid-prices.

Of course this means that at this quite early stage of “netting”, before any in-house model applies and right AFTER a mark has been performed on every position, all the parties must have agreed both on the CDS “consensus mid” prices and on the “collateral” price involved altogether. This means also that everyday any CDS price difference is reconciled along with any “collateral” price difference so that the totals match in full. This is real money here transiting from one bank to the other way before any material “event” occurs like a bankruptcy or a restructuring or else. This has a real direct impact on every bank’s revenue line as the quarter runs. No party would accept to be owed some money or just on a very short while like a day or two. What is certain behind this footnote here is that counterparties could reconcile CDS prices differences against the revaluation of the assets pledged as “collateral” from the past and for the future. Thus no price difference can escape the scrutiny of this “collateral netting” stage based on the market counterparties views. The reference to “legally enforceable master netting agreement” points to the fact that if a party reneges at agreeing on a consensual valuation for CDS prices then it is violating the law, no less. Almost all the trades and positions contained in the “tranche book” of CIO where in that situation, ie under the regime of “legally enforceable netting agreements”. And this breach of law is “enforceable” like the day after.

This reconciliation is thus pretty mandatory, and is done daily for indices and related tranches. It matters really to repeat that all (except for really few small exceptions) the CDS trades of the “tranche book” of CIO were governed by a “legally enforceable master netting agreement”. Thus no price difference could survive more than a day and more than 99% of them would spark adjustments through the “collateral revaluation” that was mandatory through the netting process. Thus those price differences at CIO London could not have been missed for longer than one day, had the control functions at JpMorgan been failing all of a sudden.

Ultimate “Fair Value” for credit derivatives

It remains to see how the risk models at Jp Morgan process this performance attribution based on the same “underlying instrument”. This is when the “assessed” values for “basis risks” and “skew risks” are set by the firm as per its internal models. Right through this process, quite logically, the firm ALSO determines the price uncertainty that it faces. This assessment naturally gives way to the liquidity reserve requirements and the capital provisions as needed. One can turn then to page 91 to look at how the firm discloses what it calls itself “Assets and liabilities measured at fair value on a recurring basis”. The regulators and banking industry standards have defined a protocol here too: a “fair value hierarchy” is to be quantified as per well defined rules….

Here one can see that the firm breaks down its “fair value hierarchy” between level 1, level 2 and level 3 assets. These are just 3 categories whereby “level 1” is liquid, “level 2” is “less liquid” and “level 3” is il-liquid. As per the firm policy (see the exhibits of the US Senate report disclosing the firm’s valuation procedures), the “level 2” requires a frequent re-assessment of the required liquidity provision since the positions are NOT liquid even though they are NOT plainly il-liquid. They are in between the two extremes. For credit derivatives receivables the firm discloses $114 759 million for “level 2” and $11 796 million for “level 3”. The total of “level 2” and “level 3” is as expected $126 555 million for the “receivables” on “credit derivatives” (no CDS can be deemed “level 1” ie “liquid” in fact). Likewise the corresponding reported “Total Fair Value” is $ 6 625 million. The figures are consistent, absent the liquidity reserves that should be set anyway in some other places in the ledgers. The “netting” amount is not broken down though: this suggests that liquidity provisions are computed on an aggregate basis for credit derivatives supporting the view that both level 2 and level 3 amounts require a reserve altogether.

The table on page 92 below provides some details:

Yet one cannot know how the collateral “legally enforceable” netting stage is being operated on “level 3” versus what is done for “Level 2” for example. In particular the firm does not provide the balance of “initial collateral requirement” that it receives and that it gives to its counterparties. These “initial collateral requirements” are usually considered as the “de minimus” liquidity reserves. The firm provides however here in footnote “e” some useful descriptions: “(e) As permitted under U.S. GAAP, the Firm has elected to net derivative receivables and derivative payables and the related cash collateral received and paid when a legally enforceable master netting agreement exists. For purposes of the tables above, the Firm does not reduce derivative receivables and derivative payables balances for this netting adjustment, either within or across the levels of the fair value hierarchy, as such netting is not relevant to a presentation based on the transparency of inputs to the valuation of an asset or liability. Therefore, the balances reported in the fair value hierarchy table are gross of any counterparty netting adjustments. However, if the Firm were to net such balances within level 3, the reduction in the level 3 derivative receivable and payable balances would be $10.4 billion and $11.7 billion at March 31, 2012, and December 31, 2011, respectively; this is exclusive of the netting benefit associated with cash collateral, which would further reduce the level 3 balances.”

The phrasing again is cryptic. But the bold sentences state 2 things. One is that this ultimate “fair value” here does NOT reflect “the valuation of an asset or liability” based on basic “valuation inputs” and transparency. Indeed the reserves and other fair value adjustments are missing in full. Two is that at least the “hierarchy” disclosed here in the table between level 2 and level 3 amounts do NOT reflect specific netting adjustments. Thus this ultimate ‘fair value’ reported on a “recurring basis” is NOT TRANSPARENT with regards to market prices, to “basis risks” and therefore to necessary liquidity or mandated provisions. Thus this “fair value” is NOT the “’mark to market” value defined as per the requirements set by the SEC since 1992 in its annual report or by the OCC in 1993 or by the “group of 30” that chaired by Paul Volcker and the then CEO of JpMorgan (1993 too). The breakdown of this “fair value” between level 2 and level 3 is therefore just indicative of the share of the “totally il-liquid” CDS positions in the firm versus “not liquid” other positions. Thus this “fair value” paradoxically is NOT what the counterparties of Jp morgan would see for example on their side of these very same CDS trades that they have against JpMorgan. One could have guessed that conclusion already when the bank had specified that it was NOT basing its aggregation on the “reference obligation” but on a cloud of diverse CDS contracts and “basis risks”…. However this is this “fair value” figure above that the firm reports for itself in its books and records. And the text mentioned here provides some clues as to how the bank proceeds despite the admitted lack of transparency.

Just that note here shows that the July 2012 $660 million restatement was NOT transparent structurally so in many aspects….Fundamentally, given what “mark to market” means for sure since 1993 for all regulators and the whole banking industry, this “price difference between CIO and the IB” that had lasted was just an invention since it would be cleared once at the “legally enforceable” netting, refined a second time through the collateral and margin call associated postings, and ultimately completely erased through the assessment of every single basis risk to perform the “Level1-Level2-Level3” hierarchical breakdown.

The firm ironically enough explains how to start from market prices and how, from this initial “gross fair value” (which is the step where CIO London had contributed and was overridden already through the routine enforceable netting process), it lands onto its ultimate NON TRANSPARENT net “fair value”. As it will be explained now, this “ultimate fair value” is anyway NOT the one that will be used in the earnings and the firm explains a little bit why but in a twisted way that indeed is NOT transparent once again. This simply shows that, whatever the price difference that may have lasted between CIO and the IB, it would have been overridden once by the in-house model risks in their assessment of “basis risks” and “skew risks”. And it would be overridden a second time before the firm actually set the fair value on these credit derivatives that would quite mathematically determine the earnings.

But first it matters to spend some time reading in full how the bank arrives at this ultimate ‘fair value’ through the use of mathematical risk models and performance attribution procedures that are run between the “cash” assets and the derivative exposures across the operating units of the banking group.

The first concept is the whole process labeled as “Valuation” in the 10th May 2012 10-Q annual report (page 97) and as “Fair Value Measurement” on Note 3 in the 2011 annual 10-K report. Please keep in mind here that every “basis risk” and “skew risk” are among the cases covered by the key sentence “If listed or quoted prices are not available, fair value is based on internally developed models that consider relevant transaction data:”….

See first the page 97 of the May 10th 2012 10-Q report.

“Valuation The Firm has an established and well-documented process for determining fair value. Fair value is based on quoted market prices, where available. If listed or quoted prices are not available, fair value is based on internally developed models that consider relevant transaction data such as maturity and use as inputs market- based or independently sourced market parameters. For further information on the Firm’s valuation process and a detailed discussion of the determination of fair value for individual financial instruments, see Note 3 on pages 184-198 of JPMorgan Chase’s 2011 Annual Report. For instruments classified within level 3 of the fair value hierarchy judgments used to estimate fair value may be significant. In arriving at an estimate of fair value for an instrument within level 3, management must first determine the appropriate model to use. Second, due to the lack of observability of significant inputs, management must assess all relevant empirical data in deriving valuation inputs - including, but not limited to, transaction details, yield curves, interest rates, volatilities, equity or debt prices, valuations of comparable instruments, foreign exchange rates and credit curves. Finally, management judgment must be applied to assess the appropriate level of valuation adjustments to reflect counterparty credit quality the Firm’s credit worthiness, constraints on liquidity and unobservable parameters, where relevant. The judgments made are typically affected by the type of product and its specific contractual terms, and the level of liquidity for the product or within the market as a whole. The Firm has numerous controls in place to ensure that its valuations are appropriate. An independent model review group reviews the Firm’s valuation models and approve them for use for specific products ( comment here: see the “VaR History” document on this website). All valuation models of the Firm are subject to this review process. A price verification group, independent from the risk-taking functions, ensures observable market prices and market-based parameters are used for valuation whenever possible. For those products with material parameter risk for which observable levels do not exist, an independent review of the assumptions made on pricing is performed. Additional review includes deconstruction of the model valuations for certain structured instruments into their components; benchmarking valuations, where possible, to similar products;

12 validating valuation estimates through actual cash settlement; and detailed review and explanation of recorded gains and losses, which are analyzed daily and over time.

Valuation adjustments, which are also determined by the independent price verification group, are based on established policies and applied consistently over time. Any changes to the valuation methodology are reviewed by management to confirm the changes are justified. As markets and products develop and the pricing for certain products becomes more transparent, the Firm continues to refine its valuation methodologies. Level 3 financial instruments

The following table presents the Firm’s primary level 3 financial instruments, the valuation techniques used to measure the fair value of those financial instruments, and the significant unobservable inputs and the range of values for those inputs. While the determination to classify an instrument within level 3 is based on the significance of the unobservable inputs to the overall fair value measurement, level 3 financial instruments typically include observable components (that is, components that are actively quoted and can be validated to external sources) in addition to the unobservable components. The level 1 and/or level 2 inputs are not included in the table. In addition, the Firm manages the risk of the observable components of level 3 financial instruments using securities and derivative positions that are classified within levels 1 or 2 of the fair value hierarchy. The range of values presented in the table is representative of the highest and lowest level input used to value the significant instruments within a classification. The input range does not reflect the level of input uncertainty, instead it is driven by the different underlying characteristics of the various instruments within the classification. For more information on valuation inputs and control, see Note 3 on pages 184–198 of JPMorgan Chase’s 2011 Annual Report. »

“the determination to classify an instrument within level 3 is based on the significance of the unobservable inputs to the overall fair value measurement “….That is what this is all about in the overriding stage that follows the “legally enforceable” netting stage whereby CIO itself was already overridden by the collateral teams of the IB : “Management” sets the adjustments helped here by “numerous controls” like VCG or MRG for the risk metrics. Valuation is certainly the number one risk of all risk measures. “Basis risk” is certainly the one risk on valuation that involves altogether senior management, “judgment” of legal teams and “internally developed models”. This is 1-0-1 of common sense for derivatives risks as heralded by the SEC, Volcker, JpMorgan and the OCC since 1993. This 1-0-1 of the actual danger conveyed by credit derivatives that was amply confirmed in 1998 after the demise of LTCM. But other risks must also be checked as they will definitely serve to make the performance attribution between the other assets, the other hedges, and the other trading positions in the firm as a whole. The stages are summarily described in this 10-K report above.

At this stage the valuation process of JpMorgan has been pictured. It matters now to unearth where exactly the mismarking occurred at Jp Morgan: it was done at the “ALCO stage”. The 2011 annual 10-K report provides a bit more descriptions of the whole firm process on page 184. And one will understand that this “fair value” sophisticated process does NOT produce the reported “fair value”, also called “carrying value”, as senior management will add a last minute adjustment layer to convey the very final “carrying value” ultimately in the books and records. This last adjustment is based for example upon the “fair value of the underlying collateral” for assets that are carried at “fair value” but on a “non recurring basis”. Once decoded this means that indeed senior management re-instill a dose of “basis risk” and a dose of “skew risk” here and there at the ultimate stage. This is the “ALCO stage” mentioned just before : “ Note 3 – Fair value measurement

JPMorgan Chase carries a portion of its assets and liabilities at fair value. These assets and liabilities are predominantly carried at fair value on a recurring basis. Certain assets and liabilities are carried at fair value on a nonrecurring basis, including mortgage, home equity and other loans, where the carrying value is based on the fair value of the underlying collateral.

The Firm has an established and well-documented process for determining fair values. Fair value is defined as the price that would be received to sell an asset or paid to transfer a liability in an orderly transaction between market participants at the measurement date. Fair value is based on quoted market prices, where available. If listed prices or quotes are not available, fair value is based on internally developed models that

13 consider relevant transaction data such as maturity and use as inputs, market-based or independently sourced market parameters, including but not limited to yield curves, interest rates, volatilities, equiy or debt prices, foreign exchange rates and credit curves.

Valuation adjustments may be made to ensure that financial instruments are recorded at fair value. These adjustments include amounts to reflect counterparty credit quality, the Firm’s creditworthiness, constraints on liquidity and unobservable parameters. Valuation adjustments are applied consistently over time. • Credit valuation adjustments (“CVA”) are necessary when the market price (or parameter) is not indicative of the credit quality of the counterparty. As few classes of derivative contracts are listed on an exchange, derivative positions are predominantly valued using internally developed models that use as their basis observable market parameters. An adjustment is necessary to reflect the credit quality of each derivative counterparty to arrive at fair value. The adjustment also takes into account contractual factors designed to reduce the Firm’s credit exposure to each counterparty, such as collateral and legal rights of offset. • Debit valuation adjustments (“DVA”) are taken to reflect the credit quality of the Firm in the valuation of liabilities measured at fair value. The methodology to determine the adjustment is consistent with CVA and incorporates JPMorgan Chase’s credit spread as observed through the credit default swap market. • Liquidity valuation adjustments are necessary when the Firm may not be able to observe a recent market price for a financial instrument that trades in inactive (or less active) markets or to reflect the cost of exiting larger than- normal market-size risk positions (liquidity adjustments are not taken for positions classified within level 1 of the fair value hierarchy; see below). The Firm estimates the amount of uncertainty in the initial valuation based on the degree of liquidity in the market in which the financial instrument trades and makes liquidity adjustments to the carrying value of the financial instrument. The Firm measures the liquidity adjustment based on the following factors: (1) the amount of time since the last relevant pricing point; (2) whether there was an actual trade or relevant external quote; and (3) the volatility of the principal risk component of the financial instrument. Costs to exit larger-than-normal market-size risk positions are determined based on the size of the adverse market move that is likely to occur during the period required to bring a position down to a non-concentrated level. • Unobservable parameter valuation adjustments are necessary when positions are valued using internally developed models that use as their basis unobservable parameters – that is, parameters that must be estimated and are, therefore, subject to management judgment. Unobservable parameter valuation adjustments are applied to mitigate the possibility of error and revision in the estimate of the market price provided by the model.

The Firm has numerous controls in place intended to ensure that its fair values are appropriate. An independent model review group reviews the Firm’s valuation models and approves them for use for specific products. All valuation models within the Firm are subject to this review process. A price verification group, independent from the risk-taking function, ensures observable market prices and market-based parameters are used for valuation wherever possible. For those products with material parameter risk for which observable market levels do not exist, an independent review of the assumptions made on pricing is performed. Additional review includes deconstruction of the model valuations for certain structured instruments into their components and benchmarking valuations, where possible, to similar products; validating valuation estimates through actual cash settlement; and detailed review and explanation of recorded gains and losses, which are analyzed daily and over time. Valuation adjustments, which are also determined by the independent price verification group, are based on established policies and applied consistently over time. Any changes to the valuation methodology are reviewed by management to confirm that the changes are justified. As markets and products develop and the pricing for certain products becomes more or less transparent, the Firm continues to refine its valuation methodologies.

The methods described above to estimate fair value may produce a fair value calculation that may not be indicative of net realizable value or reflective of future fair values. Furthermore, while the Firm believes its valuation methods are appropriate and consistent with other market participants, the use of different methodologies or assumptions to determine the fair value of certain financial instruments could result in a different estimate of fair value at the reporting date. Valuation Hierarchy

A three-level valuation hierarchy has been established under U.S. GAAP for disclosure of fair value measurements. The valuation hierarchy is based on the transparency of inputs to the valuation of an asset or liability as of the measurement date. The three levels are defined as follows. • Level 1 – inputs to the valuation methodology are quoted prices (unadjusted) for identical assets or liabilities in active markets. • Level 2 – inputs to the valuation methodology include quoted prices for similar assets and liabilities in active markets, and inputs that are observable for the asset or liability, either directly or indirectly, for substantially the full term of the financial instrument. • Level 3 – one or more inputs to the valuation methodology are unobservable and significant to the fair value measurement. A financial instrument’s categorization within the valuation hierarchy is based on the lowest level of input that is significant to the fair value measurement.”

Let’s now ascertain one important thing related to “marks” versus “liquidity reserves” which both are integral parts of the “mark to market” in theory: the “fair value” price is an “exit price”, not even a “mid price” as the “consensus prices” are perceived to be. For “level 2” and “level 3” or any “not perfectly liquid” CDS, the firm MUST therefore assess at senior management level which liquidity reserves are mandated to account for the difference between the “exit price” and the “consensus mid price”. But that is NOT reflected in the “fair value” figures that were inherited from the “legally enforceable” netted amounts. And it does not show thereafter in a transparent way as the firm stated. The process is non transparent and based upon “unobservable inputs”. The associated liquidity reserves are spread throughout other items but one cannot know either “how” or “how much” is set and “where”. Thus, for example, “management” uses CVA-DVA-Liquidity adjustments or other in-house developed models to arrive at the banks ultimate “fair value” that determines the accounts and the information to markets, regulators and investors. The bolded parts remind that on just all those points CIO had elevated the issues that all called for massive reserves for the “tranche book” alone. As this part shows, all those reserves and provisions were taken AFTER the “legally enforceable” netting had been done, which itself was a stage where CIO prices were mechanically overridden already. The senior management knew better in any way how those reserves had to be taken since only the net exposure of the bank would matter in the very first place.

It remains to see “how” the firm eventually reports the performance attribution once “management” and the “numerous controls” have done their job. Here it occurs that senior management chooses to adopt or not the “fair value election” as a measure for performance as explained in this extract. It is not always the same even though credit derivatives themselves are all placed under a “fair value” regime usually. There are exceptions, but not for the “tranche book” of CIO for which the process was as transparent as one can imagine in the firm.

The second concept that drives a lot of the performance attribution is called broadly the “fair value option”. There are many more considerations that “management” is having aside from “basis risk”, from “skew risk”, from “mid to exit prices”, or from “performance attribution across partnering business units”….Senior management shall also care about ”mitigate income statement volatility” (this is NOT equivalent to smoothing the earnings quarter on quarter, no…), “accounting for hybrid instruments” (typical of “credit hybrids” business at the IB that was just being transferred for the “tranche” part in late 2011), or dealing with IB related “structured notes” or other securitizations (again a direct reference to “credit hybrids” business). In short the other consideration further relate to either “markets expectations” or “client relationship” or “deconsolidated or nonconsolidated or nonrecurring” exposures. Here is the extract of this option impact on the accounts for the 2011 10-K annual report (page 198): “

“Note 4 – Fair value option

The fair value option provides an option to elect fair value as an alternative measurement for selected financial assets, financial liabilities, unrecognized firm commitments, and written loan commitments not previously carried at fair value. Elections Elections were made by the Firm to:

• Mitigate income statement volatility caused by the differences in the measurement basis of elected instruments (for example, certain instruments elected were previously accounted for on an accrual basis) while the associated risk management arrangements are accounted for on a fair value basis; • Eliminate the complexities of applying certain accounting models (e.g., hedge accounting or bifurcation accounting for hybrid instruments); and/or • Better reflect those instruments that are managed on a fair value basis. Elections include the following: • Loans purchased or originated as part of securitization warehousing activity, subject to bifurcation accounting, or managed on a fair value basis. • Securities financing arrangements with an embedded derivative and/or a maturity of greater than one year. • Owned beneficial interests in securitized financial assets that contain embedded credit derivatives, which would otherwise be required to be separately accounted for as a derivative instrument. • Certain investments that receive tax credits and other equity investments acquired as part of the Washington Mutual transaction. • Structured notes issued as part of IB’s client-driven activities. (Structured notes are financial instruments that contain embedded derivatives.) • Long-term beneficial interests issued by IB’s consolidated securitization trusts where the underlying assets are carried at fair value.”

About the “fair value” and the “carrying value” when all is said

Thus the bank does state that not all the assets or liabilities are “carried” at “fair value” even though they should. Here tens of $billion are under consideration. The order of magnitude is impressive but not so big when sized at the dimension of JpMorgan itself. The CIO investments alone weighed $350 billion or more. The bank total assets were worth more than $2 trillion. And the “particular strategy” at CIO was a strategic hedge for the firm in case of a global shortage of liquidity having the ability to generate $2 to $5 billion gains in a crisis…. Some assets or liabilities are thus not “carried at fair value” on a “recurring basis” when in particular their valuation depends upon the fair value of the collateral that is controlled by third parties, not by the bank. This is typical of “off-shored” skew exposures or similar “basis” exposures held by third parties vehicles (called VIEs generically in the 10-Q reports of JpMorgan). But the bank Jp Morgan would not make this clear in any 10-Q or 10-K report despite the $3 trillion potential exposure stated as “identical underlying” that shrinks down to almost zero after nettings.

On the face of it, the root cause for this distinction between “fair value” and “carrying value” above seems to be related to “PCI” loans ie loans that are problematic potentially. They are problematic in that their future cash flows are somewhat “uncertain”. These are assets that the firm needs to find a “credit hedge” for in general as they look “credit impaired” in jargon. As such, whichever class they had belonged to they sort of became part of “other assets” once they had been deemed “problematic”. And one should actually take a broader view upon what drives the last adjustment between the “in-house faire value” and the “ultimate carrying value” in relation to credit derivatives. One can refer to the definition on page 170 of the 10-Q published in May 2012: “Purchased credit-impaired (“PCI”) loans: Represents loans that were acquired in the Washington Mutual transaction and deemed to be credit-impaired on the acquisition date in accordance with FASB guidance. The guidance allows purchasers to aggregate credit-impaired loans acquired in the same fiscal quarter into one or more pools, provided that the loans have common risk characteristics (e.g., product type, LTV ratios, FICO scores, past due status, geographic location). A pool is then accounted for as a single asset with a single composite interest rate and an aggregate expectation of cash flows.”

Those loans require hedging and projections, therefore a lot of modeling based upon assumptions and common “underlying references”. As the case of the “London Whale” exemplifies, this process was NOT limited to PCI loans as far as the whole valuation of the firm is concerned. One may assume indeed that the “tranche book” of CIO was at least in part devoted to build some macro “credit hedges” against those many “PCI loans and the like”….. Dimon or Drew could certainly tell whether the assumption is correct and to which extent it is…. They should concur given the “DNA” of the “tranche book” that they wanted to have in 2006…

There were many other liability-related risks like “CVA”, “DVA”, VaR, RWA, long term liabilities, that required the same need to “pool” the risks, shrink them down to common “underlying instruments” so that the firm could also protect itself for the future at the minimum cost of course. One type of risk in particular was the lack of liquidity of some existing exposures that were NOT necessarily consolidated in full in the balance sheet and were NOT audited… Again one may make the same assumption as to the role that the “tranche book” of CIO was expected to play for the whole firm here. And a similar question should have been asked to Drew or Dimon or both…Unless all the investigation teams already had the answers.

The picture below shows that some derivatives trades are precisely meant to hedge those risks.

To be sure the amount $85 377 million is the one that is the “carrying value” on the “consolidated Balance Sheet” as the firm specifies here in this table. And this “consolidated Balance sheet” is the one that is the basis for disclosing the earnings. But where is this figure showing actually in the 10-Q reports and how can one reconcile the way in which this figure has impacted other business units through the modeling internal to Jp Morgan? The short answer is: one simply cannot do that exercise. To start with one must notice that to arrive at this figure here of $85 377 million, the firm proceeds IN FACT to 2 netting processes. One is ‘transparent’ and standard on the face of it if one excuses the total absence of disclosure about just ALL the existing “basis risks”. Thus it is the ‘legally enforceable one’ dealing with all the positions of the “tranche book” of CIO… The other one is much less “transparent” in fact: it is not ‘legally enforceable’, it is NOT fully consolidated, it is NOT audited, and the consolidation of which is NOT transparent anyway. But it adds to the first one to produce this figure here of $ 85 377 million.

The footnote “a” here is very useful again to understand what is going on here:” (a) As permitted under U.S. GAAP, the Firm has elected to net derivative receivables and derivative payables and the related cash collateral received and paid when a legally enforceable master netting agreement exists.” Ok, this is the one operation that was referred to so far. Here one can see a bit better how this netting occurs: derivatives are netted mostly between ‘payables’ and ‘receivables’, for each CDS type separately, and a smaller fraction is further reduced with the net outstanding amounts of collateral routinely pledged or received, CDS type per CDS type. This ratio between $1 483 439 million versus $ 73 635 million show that counterparties depend mostly at 95% on CDS price accuracy on each type of CDS contract to net their huge outstanding positions one with each other. One can see that the revaluation of collateral bears only upon 5% or so of the gross notional amount at risk. There really was very little room for complacency from market counterparties about the consensus mid prices, especially on “level 2”. Of course such accuracy is impossible to achieve for Level 3 assets which are il- liquid or even for ‘really less liquid’ level 2 assets like the IG9 index, the IG14-15-16 indices, the Itraxx Main S9 index, the financial-SUB indices, the HY 8-9-10-11-14-15-16 series, and all the tranches referring to these indices when available. Here about 80% of the total outstanding exposures of the “tranche book” of CIO have just been listed… Thus about 80% of the tranche book” notional amounts were il-liquid or close. They were NOT in Level 3 though, even if Hogan brought up the idea in mid to late April 2012. As to the Level 3 assets, they were over-collateralized or subject to additional liquidity reserves for that reason of that they were

17 officially il-liquid. That overcollateralization should have therefore also applied to many level 2 exposures like the IG9 index ones and actually like 80% of the “tranche book” of CIO since early 2010. The IG9 index was indeed here among most of the other indices and tranches being held in the CIO ‘tranche book’ since 2007. The matter had been raised in 2008 already at CIO. And it was raised again in 2009 giving way already to a $30 million liquidity reserve for the tranche book of CIO already. The CDS markets were getting notoriously less and less liquid since 2007. And this lower and lower liquidity usually calls for liquidity reserves aside from the collateral operations anyway. This was just meant to predate the dangers due to the basic lack of accuracy in prices whether they were “mid”, “exit”, “bids” or “offers”, or other “consensus” ones.

More, the table below the top one shows that other collateral sources do exist but are NOT factored in the ‘fair value’ that is also the “carrying value” as the firm states here in some instances. Here is the lack of transparency showing about the valuation process. The footnotes “b” and “c” provide details.

“(b) Represents cash collateral received and paid that is not subject to a legally enforceable master netting agreement, and liquid securities collateral held and transferred. (c) Represents liquid securities and cash collateral held and transferred at the initiation of derivative transactions, which is available as security against potential exposure that could arise should the fair value of the transactions move, as well as collateral held and transferred related to contracts that have non-daily call frequency for collateral to be posted, and collateral that the Firm or a counterparty has agreed to return but has not yet settled as of the reporting date. These amounts were not netted against the derivative receivables and payables in the tables above, because, at an individual counterparty level, the collateral exceeded the fair value exposure at both March 31, 2012, and December 31, 2011.”

If one computes quickly the net of those amounts attached to footnotes “b” and “c”, one gets to approx $9 billion of collateral. This is collateral that the firm “holds” in net and that is NOT reported on the consolidated balance sheet. However it does offset losses that may well be reported on the consolidated balance sheet. How is the bank setting the line in the sand here between what is re-consolidated and what is not? Well 10-Q report does NOT show how the bank does set this line in the sand through the ‘carrying value’ attached to the official earnings reports that will make the headline news. But it is here somewhere sitting in part outside of the “consolidated balance sheet”. The firm specifies by the way that the “consolidated balance sheet” is “unaudited”. So no one can certify that the ultimate earnings figure is or is not impacted by those collateral changes here.

Yet, even though some parts of the derivative overall performance that is bundled with the collateral operation do NOT show in the “carrying value” of the derivatives themselves, they may show at a later stage through the other running units of the firm. Therefore the very last stage of the valuation process that is determined by the senior management (ALCO stage) may or may not reflect that in the “carrying value” sliding a bit away from the ultimate “ estimated fair value” then….

On page 100, the firm discloses as per the rule SFAS 107 what is labels itself as follows: “Additional disclosures about the fair value of financial instruments that are not carried on the Consolidated Balance Sheets at fair value The following table presents the carrying values and estimated fair values at March 31, 2012, of financial assets and liabilities, excluding financial instruments which are carried at fair value on a recurring basis, and information is provided on their classification within the fair value hierarchy.”

A table follows:

In the past Jp Morgan used to provide the net effect on assets and on liabilities (since 1998 at least). The bank does not do it any longer in 2012 but one can run the computation. This last minute change is performed by no other than the senior management through what the OCC would call the ALCO. ALCO stands for Asset- Liability Committee. One can easily find a definition on the web: “Asset-Liability Committee Also called ALCO. A committee at a bank charged with ensuring the bank has enough assets to pay for its liabilities. It does this by monitoring the risk of the bank's investments as well as its capital structure. It reports to the board of directors and generally must also provide information to regulators.”

What is this “ALCO stage” here that checks how the firm has enough assets to pay for the liabilities? Well let’s imagine that a firm ends a quarter with assets value matching perfectly the liabilities value. One should really expect that. At the end of the following quarter, the prices of the liabilities will have changed. And the prices of the assets will have changed. The chance is almost nil that all prices will have moved “harmoniously” across liabilities and assets altogether. JpMorgan’s top management project gains, observe slightly different results and next wonder: “What happened in hindsight versus initial projections?”…. This is the rule of chaos somewhat for big companies, not limited to big banks… There are just too many variables at play…Thus there will be a mismatch between the new value of the assets and the new value of the liabilities. And the question shall be for the ALCO: “is there truly a shortfall here or there, or should the assets be under/over stated, or should the liabilities be under/over stated?” This is likely the moment when the “other collateral” postings attached to derivatives exposures that are NOT “legally enforceable” are going to be taken in consideration. Other factors come into play as well like projections of future earnings and subsequent “tax deferred (benefits)/expense”. Any last ALCO adjustment will impact as much the ultimate earnings figure from the gross first estimate. The adjustment will be an “all-in” figure for which the determination will be opaque and actually undisclosed as such. Here as of May 10th 2012 Jp morgan at the ALCO stage overstated the value of its assets by $0.3 billion and understated the value of its liabilities by some $2.5 billion. This change made the assets value match with the liability value but it also inflated the net “carrying value” increase on the first quarter of 2012 by a total $2.8 billion. Did these ALCO members, “reporting to the board” and “informing regulators”, pick the $2.8 billion gain based on their personal assessment of the over-collateralization being worth of $9 billion and being unconsolidated and unaudited? Non-one can tell apart from the watchdogs and the ALCO members themselves. The only “consolidated balance sheets (unaudited)” that investors can see do NOT display the derivatives fair value of theses “unaudited” exposures be that the ‘ultimate estimated Fair Value’ or the “carrying” one. Thus no one can reconcile this.

But still the firm discloses its final fair value for derivatives once all the adjustments have been made on “unobservable inputs” like the ones that enter in the valuation of level 3 positions (see pages 94 and 95). The firm explains what those numbers may mean:

“Changes in level 3 recurring fair value measurements. The following tables include a roll-forward of the Consolidated Balance Sheet amounts (including changes in fair value) for financial instruments classified by

19 the Firm within level 3 of the fair value hierarchy for the three months ended March 31, 2012 and 2011. When a determination is made to classify a financial instrument within level 3, the determination is based on the significance of the unobservable parameters to the overall fair value measurement. However, level 3 financial instruments typically include, in addition to the unobservable or level 3 components, observable components (that is, components that are actively quoted and can be validated to external sources); accordingly, the gains and losses in the table below include changes in fair value due in part to observable factors that are part of the valuation methodology. Also, the Firm risk-manages the observable components of level 3 financial instruments using securities and derivative positions that are classified within level 1 or 2 of the fair value hierarchy; as these level 1 and level 2 risk management instruments are not included below, the gains or losses in the following tables do not reflect the effect of the Firm’s risk management activities related to such level 3 instruments.”

The table bellow thus gives what is the ultimate number once the final ‘level 3’ adjustments are made based on the Level 1 and level 2 performances that are already factored in.

Thus the “unobservable inputs” relating to “level 3” apparently changed the ‘fair value’ on ‘receivables’ from $6 625 million down to $4 808 million. Still one will not see “where” it shows in the “consolidated balance sheets (unaudited)” or how it impacts the other assets or liabilities…..or what the level 3 related hedging costs were

The picture is quite incomplete. As explained some pieces of the jigsaw are found missing: “fair value election”, “fair value option”, SFAS107… Yet it is enough to recall the chain of key words: “identical underlying”, “gross receivables-gross payables-legally enforceable netting”, “level 3 analysis-fair value adjustments- fair value election”, “ALCO-Fair value option-ultimate Fair Value-Carrying value-Other collateral”. One will simply have to understand that the process is NOT transparent if only because the omnipresent “basis risk” is NOT disclosed at any stage, is NOT audited, is NOT fully consolidated, and is NOT reported as such although it impacts $3 trillion of notional amounts.

Why is the emphasis set on the “basis risk” then? Why is this document setting the focus on a valuation process where clearly the “basis risk” is central, undisclosed and opaquely spread throughout the process? Because the

“tranche book” at CIO that will be known as the “SCP” in the “London whale” scandal has been designed, deployed, adjusted all along since 2006 by Dimon and Drew with a view to protect the bank against the “basis risk”. This series of observations was made to emphasize that the valuation of this “tranche book” of CIO was ALSO spread across just all the stages of the valuation process of Jp Morgan. It therefore could not be the case that the quite early “estimate P&L prices” communicated by CIO London were determining the books and records of the firm at any stage, even the earliest one. To be sure here, even the initial prices coming from the “estimate P&L” would be overridden if only because this “tranche book” was hedging deliberately some “basis risk” at JpMorgan. Since the “basis risk” would be assessed only after the ‘legally enforceable’ nettings and collateral postings, these “estimate P&L” prices would be overrriden by he IB collateral teams mechanically. Thereafter these prices would be erased by others if only because at the “fair value hierarchy” ‘(level 3) stage the tranche book position may have an overlapping mark to market price for the firm Level3 risks. And even then, since some unaudited, “eliminated in consolidation” positions would be re-consolidated at the “ALCO stage” these prices would be overridden again, and maybe again for the SFAS 107 purpose.

The bank never provided the needed transparency on this intimate connection here between the “tranche book” and the “$3 trillion basis risk” even through the restatements of August 2012. It is sad as this would have showed how this “tranche book’” at CIO operated as a hedge in the firm. This is a big miss that is “strategic”. This miss alone will explain why later in this document the restatement of August 2012 should not be trusted anyway. This is the root cause for the inconsistencies that will be described later about the restated figures themselves. The parts below that are not developed here would only document further that the firm had deployed a genuine infrastructure about this “basis risk” that CIO would manage for a firm-wide quite strategic hedging purpose.

i. What is audited and what is not audited 1. Main reference tables a. GCB, DCM, CIO, treasury, liquidity reserve i. Mark to market, Measuring Fair value and fair value election

2- What is the actual bonanza brought up by the “London whale” event? a. Clearing the legend: Table on gross P&L balances and Tangible equity gains

How to reconstruct at best the path of the performance of the credit derivatives at Jp Morgan?

As the part on the valuation process showed through the words ‘identical underlying’, Gross amounts, netted amounts, ‘estimated fair value’, ‘carrying value’ ” the valuation process at the firm follows grossly 4 stages. First there is a mark consensual selection (CIO London is already overridden at this stage) and a stringent penny pinching reconciliation. Second there is a collateral netting process involving market players and first liquidity reserves. Third there is a performance allocation process run through the use of ‘identical underlying’ references in risk terms. And fourth there is a final ALCO stage were the final ‘ estimated fair value’ is switched into a “carrying value” by the senior management alone. At the last stage the ALCO bundles altogether “elections”, “options” and a last adjustment visible through the SFAS 107 reporting rule. Regulators and the board of directors are informed of the ALCO adjustments by mandate. Investors are NOT informed in a granular way as it was shown, ie they cannot reconcile the process followed by the ALCO.

It was also seen that many types of CDS, or credit derivatives in a broader wording, could exist. They may refer in part to the same ‘underlying’ risk but they would have differing prices. Those differences constitute a ‘basis risk’ on the valuation that is NOT disclosed anywhere in the reports in a transparent fashion. Thus one cannot monitor the ‘cash asset to relevant CDS’ basis or the “skew” risk existing between ‘bespoke’ transactions and index based positions for example, or between indices and their single name constituents, or else. All one can

21 know is that this ‘basis/skew’ risk bears potentially on a notional amount of around $3 trillion at Jp Morgan and on a massive netting thereafter that shrinks the $3 trillion figure down almost to zero.

Now these former remarks all supported the fact that the gross notional amounts in “receivables” and “payables” do reflect price changes to the finest details, at least at the start of the valuation process. Yet CIO London prices are already overridden at this stage. A very stringent reconciliation between the different price sources is mandated right after the price collection stage before the gross “receivables” or gross “payables” are computed. Otherwise, as shown in the first example, no valuation can be expected to be close to “accurate” and therefore no performance is close to be reliable. Missing a price difference renders the earnings report at JpMorgan themselves quite unreliable as shown with the very first “made up” example.

So better catch every little price difference on the same CDS contract inside the firm BEFORE netting all those exposures per contract type and “identical underlying”. As to the ‘enforceable’ netting that remains based on fine tuned prices, one can only rely upon the ‘netting’ amount that is provided through the “level1-level2- level3” breakdown and subsequent analysis. Yet one cannot rely upon the actual ‘net’ figure printed by the bank since this is NOT transparent and allegedly NOT a representation of all the “valuation inputs” involved. As explained the “liquidity reserves” associated to the structural difference between an “exit price” and the “consensual mid prices” are NOT displayed in the build-up of the “fair value” although they are an integral part of the valuation process. The bank explains that at least “CVA-DVA” operations deal with some of those reserves separately so. Thus one has to rely on the $126 555 figure (page 104) for the amount for gross notional ‘receivables’ and the ($119 930) figure on page 92 to best proxy what the actual price variations had on the firm’s performance. And, one will NOT know what the ‘basis’ or the ‘skew’ were in the prices used in the valuation at this stage.

This fair value measure is already opaque and deviated from the original price sources. And also one will NOT know what the valuation change was for those positions that were “eliminated in consolidation” but may have been partly re-included through the ALCO ultimate adjustment. This fair value is thus also not representative anyway of what the market price changes have induced in the actual net derivative positions of the firm as a whole. Thus one fair argument is that the bank could have made up any “price difference” it wanted on paper in July 2012. It could correct it afterwards via the ALCO stage as well. No one could check on that based on the sole 10-Q report information. But this is as good as it can be to see at least what the firm’s senior management actually “saw” then when receiving reports about market price changes and about the loss that was snowballing at CIO since January 4th 2012. One sure thing indeed is that this is what the bank stated in any case…..

If one wonders how the firm itself would “see” mark to market price changes in its own 10-Q, the actual restatement is quite a good guide as to what had apparently happened among the “legally enforceable” CDS contracts. It is key to remind that at least 99% of the “tranche book” trades at CIO were lodged in “legally enforceable” contracts. So the scrutiny about the “tranche book” of CIO can be performed to some extent as this book had NO “basis risk” and no “skew risk” as such. Again this is what the bank stated “in hindsight” as what it had “seen”. This may not be what counterparties saw individually.

Thus on page 104 of the 10-Q report disclosed on May 10th 2012, one can see the following figures:

And in the restated 10-Q as disclosed in August 2012, the firm provides figures that differed, only based on “mark differences” between CIO and the IB:

A table summarizing those differences shows below:

As one can see, the price differences only affected the “gross” figures inferred by Jp Morgan systems based on the original ‘marks’ extracted from markets. The key feature is that the ‘netting’ totals are changed as well. This is mechanical indeed. Yet as per the theory deployed by the bank, had those price differences been “unknown” in the bank, they could not have been missed by the counterparties of CIO. One can gobble that JpMorgan maybe was “complacent about CIO”, that JpMorgan was having a quite unprecedented series of control failures all at the same outside of CIO then, that JpMorgan had noticed no warnings from ICE, that JpMorgan also had carried for 20 years a valuation loophole in a standard process that JpMorgan had imposed to the regulators and the whole US Banking industry…One could gobble also that all this occurred by chance right when the “tranche book” at CIO weighing 40% of the firm’s total VaR went through dramatic losses. But one cannot imagine that the whole US banking industry and the whole world industry in Credit derivatives did miss that $660 million in price differences about what was then “one if not the biggest client” in credit indices in the world. No that tale really does not hold water. A lot of money was missing here actually on not so many positions. Artajo would state to Pinto and Macris on March 23rd 2012 that “one position” alone generated a difference of $250 million…. It was hard to miss if it had ever existed. The figure was stunning the London CIO chief and the UK CEO of JPM altogether.

Actually Pinto did not bother being assertive that Jp Morgan had NO dispute then…If such price differences had really persisted more than just one session, they should have impacted the ‘netting’ amounts that were “legally enforceable” in the eyes of the counterparties. And they should have sparked disputes in late March 2012 already. The amount was pretty significant. The price differences in question were NOT triggering disputes then….One should read again the march 23rd 2012 call between Pinto, Macris and Artajo were Pinto is assertive: they do not have disputes at the IB. And CIO was the “client” of the IB for dealing with its margin calls and collateral issues….

For the purpose of the restatement allegedly caused by lasting price differences between CIO and the IB inside JpMorgan, the “netting” figure had to change too…But the counterparties did NOT complain at all in late March 2012. This restatement story definitely does not hold water. This reconciliation table above thus shows that those price differences were internal to Jp Morgan and NOT an issue for the market counterparties having “legally enforceable master agreements” in place with CIO for the “’tranche book”. This was the case for more than 99% of the positions of the “tranche book” of CIO as said already. As to the rest of the process, any price difference would be reconciled and adjusted, ie erased. And next the process is opaque as explained before. Logically so, the other figures do not change as the next stages are based on just one price anyway. The restatement will be further commented later on. It is enough to see here that indeed price changes would alter

24 the “gross” receivable and “gross” payable figures displayed in the table that specifies the derivatives “designated as hedges” along with netting amounts naturally.

One may remember that the overall price difference was of $660 million but that the restatement was of $459 million due to “other adjustments” made in the course of the valuation process of the firm. A table on page 93 of the August 2012 restated 10-Q provides some more details. It matters to see that the $660 million due to price changes impacted in full a line called ‘principal transaction’. If “principal transaction” is meant to reflect a pure trading transaction involving the market counterparties through the day to day collateral and margin calls, this again makes no sense. This is the same remark that applied to the “netting” figures.

As a further confirmation of the very early impact of those price differences on may go to page 109 in the May 10 2012 110-Q report and page 112 for the restated 10-Q:

Here this was NOT $660 million but $692 million. Here is therefore another “faire value” change. The firm specifies here in footnote “b”: “The amounts are shown on a gross basis, before the benefit of legally enforceable master netting agreements and cash collateral received by the firm”. Yet one can see that the outstanding amounts are totally the same. It is only the early ‘fair value’ determination stage that makes a difference. Therefore the move down from $692 to $660 million impact was already the byproduct of collateral management, based on master agreements that are “legally enforceable”.

So the “gross” amounts and the “netting” amounts should provide a fair picture of what the ALCO and other senior managers saw through the “London whale” scandal. This is based on market price changes effects overall, on CIO positions, on IB positions and other positions that had been netted inside the firm and collateralized through “legally enforceable master agreements” with market counterparties. All this was based therefore upon “consensual mids”. And the change from $660 million to $692 million shows once again that operations had been performed by the IB staff handling the collateral management of the “tranche book” on behalf of CIO. And this again shows that the IB staff was hunting for every tiny price differences to be adjusted and ultimately flattened through the netting…. And they had found $32 million here….If the tale supporting the restatement is to be trusted one must conclude that the IB collateral staff had missed $660 million of price differences between April 3rd and July 12th 2012, but they “found” $32 million of additional price difference thereafter. Does it make sense?

Thus if one summarizes: the bank displays an inconsistent reporting for the restatement as the changes in “netting”, in “principal transaction” and in the “total” for the “fair value” change (table above as circled) all that indicates that the IB staff would have annihilated the price differences at the “netting stage” anyway through their routine operations and matching checks with counterparties’ claims. This is what the logic behind the firm- wide valuation process dictates. This is what the IB collateral teams did since 2006….Thus for this $660 million of price difference to have persisted over one day, it must be that the IB collateral staff did NOT perform its job as usual on behalf of the “client” CIO after March 2012 month end.

The actual path of credit derivatives performance of the CIO, of the IB and of Jp morgan

It has been shown that the “gross receivables” and “gross payables” reflected market price changes. It has been showed that the firm was not transparent on its own assessment of the “basis risk” while determining these gross amounts. However the “tranche book” positions at CIO had only indices and tranches were based on indices at more than 99%. They were all quotes independently at the early stage of the valuation process. Thus there was no “basis risk” or “skew risk” embedded in the determination of these gross amounts for this “tranche book” of CIO. A price difference of $660 million has been alleged for what would be a $5-6 billion total loss based on market prices changes. This is material (like 12% of the total) but this is not changing the big picture anyway. Thus a fair basis can be found to observe the actual impact of market price changes as seen both by Jp Morgan and its market counterparties. And one can then see what the “London whale” story really was about with regards to this alleged “credit derivatives trading loss”….

There is not a single doubt that CIO suffered a multi-billion loss based on credit indices and related tranches. All that loss came from quoted prices in the markets and changes on “consensus mid prices” anyway. There is not a single doubt as well that, being under “legally enforceable” netting agreements, the bank had to agree day after day with its counterparties. There is therefore NO doubt that, as of March 30th 2012 or April 3rd 2012, there was NO dispute between Jp Morgan and its counterparties YET. BUT the bank “in hindsight” in July 2012 would state a cause for dispute worth of $600 million or more as of march 30th 2012. The snag here is that dispute here was allegedly visible as of March 30th 2012 or April 3rd 2012… “in hindsight” if the restatement tale is to be trusted. It does not have to be trusted anyway for what follows…Yet it matters to come back to this restatement one more time as it will show a bit more of what the collateral management entails on a day to day routine.

Is it simply possible that just ALL the market players silenced this $660 million dispute then on March 30th 2012 against JpMorgan CIO? No, it is NOT possible. The magnitudes at stake as mentioned in the first part were too big. More, there was actually NO risk that such a $660 million dispute on collateral netting and subsequent valuation could exist within Jp Morgan because of CIO London estimate P&L prices. The IB itself would have disputed CIO differences if it had had to. It matters to remind that the CIO had open trades on almost all the products in direct with the IB of Jp Morgan. The IB would NOT have to dispute CIO’s prices actually, even in April 2012. The IB was actually managing the margin calls for CIO with IB’s prices as Pinto pointed out on this crucial call of March 23rd 2012. It matters to remind also that almost all of CIO trades in credit indices were cleared through ICE which acted as a clearing counterparty here between CIO and the rest of the world, including the IB itself. IN any case CIO did not act inside JpMorgan as a level playing field contender to the IB in the mark to market process.

It matters for that purpose to remind also that CIO did NOT follow the industry standards and practices since 2007 and knowingly so. Namely CIO had no closing time for its “tranche book” and CIO did NOT adjust the tranche marks to the reference index closing price on the day. It was what the firm wanted CIO to do. Thus CIO did NOT provide marks that allowed a proper collateral management and margin call processing. CIO did NOT follow a process that would ever comply with some of the most basic US GAAP standards.

All this standard “mark to market” on CIO exposures for the “tranche book” was actually performed every day by the IB team that was dedicated to collateral management on behalf of the bank clients. The CIO was treated internally as a client of the bank Jp Morgan through the IB. Therefore price differences with the CIO were not only the “routine”, but they were adjusted by the IB every day through a necessary reconciliation process involving the market counterparties involving “legally enforceable master agreements”, through the IB itself and ICE. For its own sake and the bank’s sake, the IB had open trades with the “tranche book” of CIO on almost all the instruments being in use for this “tranche book” of CIO. And for any of those trades the IB and CIO had to

26 match to the $1 in P&L. Had it not been the case Jp Morgan was just another Ponzi scheme…..It is in that context that the following tables and charts will be displaying the performance of CIO, of the IB and of the firm quarter after quarter….One can really forget about missing $660 million in the figures that will come.

The first table shows the quarter-on-quarter estimate P&L of the CIO “tranche book” from the start of 2009 until Q2 2012. Its results up or down are of comparable scale to the ones of the “CVA” desk of the IB that was run by Rob O’Rehilly in 2012. It had a minor impact overall as the table shows. Both activities, ie “tranche book” at CIO and “CVA-DVA desk” at the IB, fulfilled hedging roles for the bank having similar and somewhat offsetting impacts on the firms’ revenue. Yet, things changed radically somewhere between Q4 2011 and early 2012 for the “tranche book” at CIO which took a dive indeed…

But as the next table will show, one sure thing is that the firm did NOT suffer such a dive as a whole and the compensating gains did NOT come from the CVA desk of O’Rehilly. The table below adds indeed the reported net credit derivative fair value change based on “gross” receivables/payables and “netting” amounts coming from “legally enforceable” master agreements….As circled one sees, as the bank “saw” it for Q2 2012, that the impact of credit derivatives overall was quite benign in comparison to the former quarters. There is no restatement here but there is the $4.4 billion loss at CIO embedded in it along with the gains that must have occurred elsewhere within the bank on a “legally enforceable” manner here for 99% of it.

The balancing gain at Jp Morgan of about $4 billion on credit derivatives may have been fortuitous. A loss of $385 million is the result for Q2 2012 as the bank “saw” it. That was it for the “trading loss” for the whole firm due to scrutinized price changes occurring on CDS, in a period where the bank had allegedly been plagued by the “london whale” scandal based on its allegedly “flawed positions”…. A loss of $385 million was therefore the ultimate impact on the firm’s revenue through this unwind of a CDS book that had been rumored to be “quite hard to get rid of for JpMorgan” (see Gregory Zuckerman again on the matter). As per the new version of the legend burgeoning in early June 2012, the bank had been locked up with these trades that were so visible, so targeted, and so illiquid all of a sudden. The bank was rumored in early June 2012 to have had a $5-6 billion loss already that may grow to double that as soon as the firm would really try to unwind them in the markets. None of that occurred.

This loss of $385 million showed in Q2 when the bank had stated that it would NOT unwind anything more in the markets. So that was it. Dimon had “moved forward”. The bank revenue would have a “hit” of $385 million over the scandal surrounding the now dismantled “SCP” that had weighed a good 40% of the firm-wide VaR for years using the most toxic derivatives in man’s memory. What a small wind down cost it was for the bank as a whole while the “tranche book” of CIO itself was slaughtered....

One may doubt the $385 million figure thinking that there may have been a loss elsewhere that was NOT reported in “credit derivatives”. Well the bank made not a single mention of that when it “explained” the miraculous $5 billion earnings for Q2 2012. Not a single regulator would raise a point of that nature. Instead they all loudly complained about the fact that bank never provided the “assets” that the “tranche book” of CIO was supposed to hedge. Market players observed that the bank had unwound almost nothing of the “tranche book” of CIO actually. All the risks remained in the “Jp Morgan universe” which included Blue Mountain then. There is NO miracle here….So, if one accepts the very simple idea that the $6 billion loss at CIO was caused by price changes, then the ultimate “trading loss” for Jpmorgan in Q2 2012 was NOT $4 billion but less than $400 million at the worst. There is NO explanation provided by the bank or any regulator about the source of the other $4 billion of gains that occurred on credit derivatives in Q2 2012 right against the losses of CIO “inside JpMorgan”. There is no clue right? Unless there was a “hedge” for the “hedge” and unless this “hedge of the hedge” had been long prepared for this panicky dismantlement necessarily so… This “hedge of the hedge” could only be deployed by the IB. And, given that CIO notoriously traded “massively” over Q1 2012, it had to be the case that the IB was already dynamically “hedging the hedge “ quite systematically no later than January the 1st 2012. Thus the IB was very, very, very, very involved day to day 4 months before the seminal articles on the “London whale”….

The bank and every regulator would claim that the IB was NOT involved in that kind of “hedge of the hedge”. But no one has to believe what the bank and all the regulators entertained on this line. There was NO such thing as a “massive loss on massive trades at JpMorgan” that may have destroyed the bank from CIO and its “tranche book”….The fact is that the bank did NOT unwind with external market players and printed just $385 million loss on the second quarter 2012 while stating that he had “got rid” of the “tranche book” of CIO. This number alone was quite “business as usual” in fact too.

One can argue that there were 2 consecutive quarters here of losses respectively of -$1 171 million and -$385 million in Q2 2012. Well, if the bank had had this loss in Q1 2012 because it had prepared to “hedge the hedge” then it had to be done by the IB and the bank had to say it in early April 2012. If such was the case, the bank said none of this, and therefore the markets were totally misled then. Thus many class actions should be started against Jp Morgan. It would have been such a naïve “mistake” on the part of the bank. But one would wonder still why the emphasis was set so much “in hindsight” by Dimon on the “second quarter” all along. It is the first quarter of 2012 that deserves a better scrutiny maybe. And one can see that this kind of outcome of -$1 171 million plus $ 385 million would be anyway completely standard when one looks at the figures spanning between Q2 2010 and Q3 2011 for example. It is even very benign if the year 2009 is a guide….

So, at this stage of the valuation (price and collateral where CIO-London inputs were already superseded by IB staff) things had remained pretty “normal” if not pretty nice for the firm actually. This is at odds with any description that the firm itself would make of the event, emphasizing the “errors” of CIO and the “control failings etc…”. This is even more at odds with the authorities’ subsequent focus. This $385 million figure strongly suggests that this was NOT such a derivative trading loss at Jp Morgan. But one may argue that, the “level 3” factor for those highly concentrated and il-liquid positions may have played a disguised role in the backyard, a kind of obscure and impenetrable role due to “high complexity”. Why not after all?

Most of the valuation process was not done yet indeed. May be that it was not the CIO “tranche book” that was at loss so much but the assets or risks that it was expected to protect. As explained before, post the “consensus price” fixing and netting process, post the collateral adjustments, there was the “risk modeling” non transparent input where the firm confidentially based its performance attribution from “identical underlying” frameworks. Here the firm centrally assessed the price of the many “basis risks” and “skew risk” that it had to estimate at the very best given the massive magnitudes at play on credit derivatives (remember $3 trillion of notional amounts

28 being shrunk to almost zero quarter after a quarter). And, next, the ALCO would anyway place its own multi- billion in $ “value” for the final SFAS 107 related adjustment. Thus the $385 million “transparent” result had every chance to be massively altered when the firm would arrive at its “carrying value” for credit derivatives. But again, one has to rely on what the bank stated it “saw” and what the bank stated in general. And as far as the bank’s own report are to be trusted, there was just that $385 million loss for Q2 2012. That result was based on credit derivatives given “mark to market” prices and subsequent margin calls and ensuing collateral adjustments as disclosed. So much for the “trading loss at Jp Morgan on credit derivatives”. By the way what was the self- assessment of the ALCO on this pure “credit derivative” performance at the firm for that matter?

The table below will show that indeed the “level 3” element and especially the assessment of the ALCO did play a role. Indeed the ALCO stage may have worsened the picture for credit derivatives. But to which extent exactly did the ALCO do that and for which period exactly?

Well the ALCO itself saw a pretty benign event for Q2 2012 as far as credit derivatives prices were concerned after all had been reviewed. Thus, IF the “tranche book” of CIO had been just a sort of “hidden prop trading hedge fund shop”, its $6 billion “trading loss” was NOT at all a “trading loss” for Jp Morgan. To say the least the loss of CIO in question had been “hedged” in advance by the IB staff no later than the very first days of Q1 2012 and this at very low cost anyway. Since even omitting that “$6 billion detail” would have been a gross fraud on the part of Jp Morgan given its “statements and writings” of the time. So one must assume that the IB – if it ever did- had “hedged” the “tranche book” of CIO by the end of 2011 already. Thus actually that was no such “hedging of the hedge” strategy run at the IB. So unless the bank lied grossly all along, this “tranche book” of CIO NEVER was a “prop trading hidden hedge fund shop” as such. This is what the $385 million versus the - $4 400 figure say.

Ironically the ALCO saw a much larger loss for Q1 2012 and an even lower loss for Q1 2012. Only -$360 million was in their cards from a $385 million “gross” direct negative performance. One can see from the former records that the figures were not altered so much in fact in general. Once again, as the table above shows, there would be some noticeable impacts on “Reported Credit FV chge” though for Q4 2009, Q1 2010,Q3-Q4 2010 and maybe Q1 2012. Few $ billions were at stake potentially which is NOT surprising when one recalls that the overall value of $1bp of “basis risk” at Jp Morgan is worth about $1.5 billion. More, the ALCO had mostly “seen” a $4 billion loss spanning between Q4 2011 and Q1 2012 that followed actually a $4 billion gain in Q3 2011….Or was this $4 loss cumulated in Q4 2011 and Q1 2012 the mere anticipation of the $4 loss that would pop at CIO in Q2 2012? It could be both….This gain of $4 billion in Q3 2011 showed up right before ‘credit hybrids’ would be shut down and before price differences would emerge between CIO and the IB day after at the very first stage of the valuation process of the firm. This $3-4 billion was equivalent to just 2-3 bps for the whole $3 trillion basis risk of the bank if one uses the first example. As such this was not unexpected to have this kind of uncertainty at times. One can see that the losses of Q4 2011-Q1 2012 were not uncommon as well since they had shown between Q3 2010 and Q1 2011, just as if there was a seasonal effect.

From those results, assuming that the “IB” is the main other trading part at JPMorgan on credit derivatives, one can therefore safely proxy what the P&L for the “IB universe” was simultaneously a gain offsetting the CIO

29 losses at the time. And one will see in the table below that the bank likely had offsetting positions, mostly controlled by the IB:

The result is shown in a dark color to picture the fact that this massive gain occurred in the shadow if the “london whale” legend most likely in the “IB universe”. The “IB universe” to be sure covers the IB as consolidated on the “consolidated balance sheet” plus the rest of the position that the IB managed… It has been constantly overlooked since then by every public investigation report. But the inference here leaves little room for the doubt as the IB was the only one large entity trading on CDS at JpMorgan. One may remember the March 23rd call between Pinto, Macris and Artajo where Pinto stated that the IB had “hedged” CIO on HY in 2011. The IB via Pinto the JPMorgan UK CEO had been mandated to “hedge CIO on HY”. One wonders “who” could have done that between Dimon or the watchdogs watching….

So it is safe to assume that the IB exposures compensated almost fully the CIO losses day after day all along the first and second quarter of 2012. That was not such a big deal at the IB which was historically a much larger player than the “tranche book” of CIO (3 to 4 times bigger overall it was said internally). The past records from 2009 show that the IB had much larger swings in P&L than the “CIO tranche book” had. The IB was making profits in 2012 and likely could have gone even bigger if needed in front of CIO as the historical record tells.

One may argue though that this is just an inference, that such a benign outcome on credit derivatives in Q2 2012, ranging between -$385 and-$360 million, may have resulted from fortunate and timely gains coming from other derivatives clusters. The last table provides a broader picture including the other related derivatives (rates and equity), all the derivatives and the earnings:

One can usefully look at the Dark blue line and yellow font labeled “Credit –rates-Equity” which is the natural cluster of derivatives for a “tranche book” like the CIO had. Here one sees an outstanding profit for Q2 2012 of $1 713 million. The former records indicate that it was a rather small profit in a series of highly volatile performances, positive or negative. It thus feels that in the context of the scandal, things were actually pretty well balanced on derivatives and just marginally profitable. This picture here goes straight opposite to what the bank would keep entertaining during Q2 2012: the bank was actually much better balance that it had been in the past! Was this a coincidence?

The other lines about the reported fair value including other derivatives show that, irrespective of the hedging strategy that the bank had, the “London whale” event looks pretty benign no matter whether one only looks at market price changes, or one includes the risk model re-allocation of the firm. Now if one looks at the line “measured earnings from actual Fair value change” and looks at the line right below indicating the “reported earnings” one sees again that the first quarter of 2012 was maybe a problem. But Q2 2012 stood out again as pretty profitable on derivatives overall.

The historical record seems to indicate that. But the SFAS 107 rule may suggest otherwise. What happened in Q2 2012 in that regard too? As suggested before, the more one looks at Q1 2012 instead of Q2 2012, the more one senses that there was an issue for Q1 2012. As such the official blame of mismarking bore on Q1 2012 and corroborates the suspicion. This would induce to say that the “London whale” scandal looks like quite a catharsis for JpMorgan…. Just looking at the very surface of those numbers….

As mentioned before, the ALCO for Q1 2012 inflated the “carrying value” by $2.8 billion while it had understated this same “carrying value” by $2.2 billion in Q4 2011. Thus the stated earnings of close to $5 billion in Q1 2012 incorporated a total $5 billion positive adjustment made by the ALCO from Q4 2011 to Q1 2012…That’s quite an optimistic stance and not a “tempest in a teapot” at all as such for any watchdog. The regulators should have been all concerned AT THE TIME ie as per April 3rd 2012 and even more so as per April 13th 2012 when this $5 billion ALCO tweak goes public as a “tempest in a teapot”. The regulators would scrutinize the price differences on these $3 trillion of basis risk, for sure.

From that standpoint, the articles on the “London Whale” can be perceived as a distraction or even a diversion from the real issue here. The board was aware and regulators were involved. So when Jamie Dimon stated “I agree with you it is a tempest in a teapot” it may have sounded like a provocation actually, and maybe an invitation to dig in the books and records of the bank. The CEO and board chairman knew what he was doing. When the authorities realized that some price differences worth of $300 to $600 million had NOT been adjusted as they should have been, they became suspicious as anyone would be. If one doubts that they had been told of those differences no later than April 13th 2012, it is enough to remember the Fed CCAR request of late December 2011 and the March 23rd 2012 call between Pinto the UK CEO of JPM, Macris and Artajo. The regulators already worried about price uncertainty, unwind costs, “sufficient consideration for concentration risk”, “sufficient documentation of the hierarchy among the prices sources” for the ‘tranche book’ of CIO and its index positions. But they would NOT need at all to talk to the London CIO Front Office “traders”. They wanted to talk to “management”….They too knew well what they were doing: they had no “trader” to hang yet.

The $5 billion 100% ALCO-led profit that induced a $5 billion reported earnings for Q1 2012 echoed an ongoing concern for regulators that had existed since Dimon had been the CEO of JpMorgan. As shown in this table, Jp Morgan had a history of placing last minute ALCO adjustments that presented a smoother earnings path. How appropriate was it for Q1 2012 no later than April 13th 2012? When one looks now at the $8.8 billion gain on derivatives for Q2 2012, one must ALSO wonder how accurate the contrite statements of August and July 2012 were actually…The bank was making big money it seems rather than avoid a cataclysm by a hair.

Regulators all alleged a high degree of unawareness in the aftermaths of the scandal before May 10th 2012. But they also clamored that next that had had quite a stringent look to all this mess right? There must have been one question that they were asking themselves: “how much was the bank making actually in Q2 2012 versus what it likely lost in Q1 2012?” One has to recourse to the concept of “tangible equity” or “hard capital” to get the full answer that they would find for themselves…..

The answer was plain visible for all the watchdogs to see: Q2 2012 looked damn good despite the losses of CIO. They must have thought “thanks” to the huge losses of the “hedge that CIO was managing on behalf of the firm since 2007”. And as the $385 million figure testifies: it looked too good to be true as of April 3rd 2012 since Q1 2012 earnings looked too rosy. It would be scrutinized for sure. Yet just none of the regulators and none of the bank top executives meeting so often together on the matter during Q2 2012 would try to talk to Iksil. This must have been fully unnecessary. But they would ALL try to meet with Iksil in order to build a “version of facts” that would turn out to be flawed here or there starting in the end of June 2012.

The complete answer about the actual “gains” made by JPMorgan between Q1 2012 and Q3 2012 was a matter of focus. The coming concept of “tangible equity” is central as it connects directly Dimon’s “wind down” plans, the programmed fate of the “tranche book” and Iksil altogether, the “basis risks” and the “skew risks”, and the market manipulation that led to the media manipulation.

How to define the ‘tangible equity’ based on Jp Morgan reports?

The concept of the ‘tangible equity’ or ‘tangible capital’ is based upon the consolidated balance sheet and what is commonly understood to be the “Hard capital” of the bank. Yes there is somehow a “soft capital” and a sort of “brick and mortar like” capital… The capital of a bank is customarily observed through the lens of the share price. But the share price can vary a lot, fast, and usually people struggle to understand these changes. Yet these changes matter a lot for a bank. Indeed, as the share price rises, the bank is perceived to have more capital and vice versa if the share price falls, “all things being held equal”, the bank is perceived to have less capital. Some banks do not have publicly quoted shares and/or have other sources of “capital”. The “capital” more generally is the cushion of money that a bank has in order to foot any last minute bills be that a sudden catastrophe or be that a market fall. It is therefore the bank’s survival line, a concept that is central in the regulators’ daily concerns and actions.

It is a well known fact that the traded share price is so volatile that it is a poor measure of the capital that a bank has on a recurring basis. One had rather look at the balance sheet to assess a more stable and rational measure of this capital. The useful figure to look for is actually first inferred through a computation that is run across the consolidated balance sheet as the difference between the total asset value and the total “debt” which carries payment claims that are senior to the shareholders’ rights. As such this is the view of “senior management” anyway in that only the “consolidated” part are taken into account as per the ALCO adjustments. In the case of JpMorgan it is unavoidable to rely on the consolidation as per senior management. The notion of consolidation applies to almost all complex banks. Indeed big banks like Jp Morgan gather many sorts of businesses, some being fully ingrained in the bank’s historic activity, some being less so with as many degrees of dependence as one can guess. As to JpMorgan, the big complex firm thus has to assess all those degrees and make a sort of “weighted sum” of all these businesses that gravitate around the brand “JpMorgan”. This is what the concept of Consolidation covers. This is by the means of this consolidation that regulators have to check whether the bank has sufficient capital. The watchdogs though have much more confidential information than the public.

And things are always more complex than they should be. It may occur that some “non consolidated” or “eliminated in consolidation” exposures require some capital to be set aside by the bank still. This is in general

32 determined by the supervising regulators as one can guess. Typically an “off-shoring” operation like the one that Bacon commanded for Dimon in March 2012 for the “tranche book” of CIO was a de-consolidating operation that required prior approval of regulators and knowingly impacted the capital of the bank. What was the impact? All the computation of the minimum capital “required” was based in those days on commonly approved standards called Basel III or Basel 2.5. These standards are in constant motion and the adoption is left to the goodwill of every bank and regulator, still subject to prevailing sovereign legislation. Thus the short answer to the question “what was the impact?”, is “well, it must be negotiated…” And then the question becomes “which would be the criteria driving the negotiation?”…. The “RWA” or Risk Weighted Asset is the basis that is used in fine to determine the capital that is required as a “minimum” and that therefore allows investors to infer the capital that is “in excess”. But this is just a computation based upon highly confidential data that investors cannot check or reconcile with facts fully. In some respects this is quite a speculative number for such a critical negotiation.

Thus the actual “capital” figure as computed form Basel-xxx standards is not really a guide for shareholders. Take the argument of “confidentiality”, or the argument of “changing standards” or the argument of “model risk”, the public need a self consistent framework to assess the capital of the bank. The investors need a much more tangible basis. It matters because if a share price publicly falls this is unquestionably the sign that the bank capital is in danger of vanishing whatever the RWA figures would suggest under any Basel XXX reference standard, whatever regulators think. Should that share price fall persisted then the bank would be soon rumored to go bankrupt. “Facts are stubborn” at times… Usually this situation of a regular fall pressures the share price even lower. And so goes a lethal spiral as one could see in 2008 that acts as a virus almost across the whole financial planet. On the opposite if the share price goes up steadily, then the bank is perceived to be safer and actually more profitable for shareholders looking forward. This supports the share price further up. And at times banks end up with largely overvalued share price. Who is fooled by that tale of “trees grow to the skies” then? Nobody is fooled for long. The subsequent correction on the way down is always problematic as explained right before.

Thus a tangible basis for the capital is crucial because it sets a sort of floor for the share price when things get bad. This is what the concept of ‘tangible equity’ is all about. This is also a consideration that drove Dimon in “his” wind down plan that targeted the “tranche book” of CIO since late 2010 and would only come to fruition on Q2 2012. As “Jamie” was reported saying all along back in 2010: he wanted to “give money back to shareholders”…. Given the information available in the public reports or even the one available in the confidential reports that only regulators have, the picture is quite incomplete. The market players can only speculate still upon what this “tangible capital” basis is worth. Here is a little reasoning that has had likely great success after August 2012 when people could at last read through the “London whale” tea leaves in the bank’s accounting ledgers just a little better…

The shareholders’ claims are considered a liability to be repaid by the bank at one point in time. If the bank goes bankrupt, they are notoriously expected to get peanuts. And if the bank survives, they strongly hope to be well rewarded for that reason above. And, as long as the bank is alive, the shareholders must assess what is the value of their claim on the assets sitting on the balance sheet with a view on what their “floor” is in share price terms. What they know is that the depositors, the employees and the debt-holders of Jp Morgan would have a priority in their own claims if the bank went in dire straits. These claims holders would access all tangible assets that were left to repay their own money in priority. So looking at the balance sheet, shareholders know that what they should rely on is what is left for them after theses senior claims had been honored, ie “intangible assets” and the rest of the “tangible assets” that are not quite easily identified in general.

To arrive at what is tangible for them the shareholders should look first at what is labeled as the “common shareholder’s equity”, divide by the number of shares, and arrive at what is called the “book value”. This share price here labeled as “book value” is NOT the one that is publicly quoted. This is the value that the bank itself sees on its consolidated but unaudited balance sheet for the claims that shareholders own against the bank. They may very well consider that the traded share price is worth more if they anticipate that the profits are going to grow be that because the bank is going to grow on its own or because the economy is going to grow or because the markets are going to go higher anyway. But this is here their speculation mostly, not a tangible basis for a long-term wise investment.

In any case, before they may speculate freely, they need to know in the first place whether this ‘book value’ is solidly established. They thus need first to remove any “intangible” part as this is quite intangible by definition. Thus they would simply deduct the total ‘intangibles assets’ from the “common shareholder equity” as stated by the bank to arrive at the ‘tangible capital’ that they can rely on. Over time they will monitor whether their reference ‘book value’ moves up because there is more and more ‘intangible assets’ or whether the bank generates “brick and mortar” capital on its balance sheet through the profits it is announcing. They thus can assess the intrinsic riskiness of their own investment or own speculation on the actual quoted share price based on the bank’s stated performance.

Now if investors observe the quoted share price trading below the ‘book value’ as reported by the bank senior management, they may start speculating that a lot of people are questioning the reliability of the profits stated by the bank irrespective of how high they might look on paper. This is a circular reasoning that finds no real limit. This divergence often announces a coming demise if history is a guide. But this is not always verified…. Regulators watch quietly that pattern when it shows up and worry. So the investors (and the regulators alike) need to scrutinize further what kind of assets their own ‘tangible equity’ relies upon. Taking the stand of depositors, debt-holders or other prioritized stakeholders, common shareholders would assume that they would be left mostly with what is labeled by the bank as “other assets”. Why is that? Because the well known asset categories like, “cash”, “securities”, “receivables”, “private equity”, “loans” go first to those other stakeholders in emergency cases. They are usually well identified because there is a known market for them. Either they can be sold out or restructured towards other participants of this well identified market. Thus common shareholders would assume that they should rely first upon “other assets” which typically are assets that not well identified in general. The cause is simple: there is no real market grouping regular buyers and regular sellers. Comparing here their tangible equity with those ‘other assets’ they would know by deduction whether they can expect to access to more classical asset classes.

It turns out that those “other assets” at Jp morgan are ultimately equivalent to “margin loans” or other atypical receivables based on quite special derivative exposures. The historical record shows that those “other assets” ballooned as Jp morgan took over Bear Stearns, Washington Mutual and other assets from AIG, SIGMA or Lehman. One can safely assume that for these “other assets” there is either no deep liquid market or no reliable price or both or worse. Otherwise these “other assets” would bear a well defined label based upon a well identified market as is the case for just all the other well defined assets classes. The reference to 2008 and the sudden growth of “other assets” may just be a sort of “anecdotal” proof but the recent history supports well the description of what “other assets” are as consolidated on the bank’s balance sheet. In any event that heavy presence of all these toxic and il-liquid “receivables” coming from Bear Stearns or WAMU or Lehman or else in the balance sheet of JpMorgan was a concern for both shareholders and regulators starting in 2008 at the latest.

If the tangible equity turns out to be larger than the value of “other assets” then one should assume that common shareholders should rely too upon the “private equity” investments of the bank or some other risky leveraged loans or some even safer assets. It would mean in other words that the shares became more of a “senior claim”

34 on the company. This would constitute quite a strong incentive into buying the share at higher price looking forward if at this moment the bank prints high profits.

Thus a strong rational basis for shareholders to pick the jp Morgan shares at its current quoted price is achieved by using the ‘book value’, the ‘intangible assets’, the resulting ‘tangible equity’ and its difference with the value of ‘other assets’. Next some finer analysis could be processed depending on how high the traded price is relative to the book value. But as explained this would bring up much less of a tangible conclusion.

So this is the frame that is going to be used to show that the bank made a bonanza gain right in 2012 throughout the event of the ‘London Whale’. It has been shown already in the former section that this event was NOT “a trading loss” for the bank, since the CIO losses were instantly compensated especially during the second quarter of 2012. There was no such “hedge of the hedge” as the “tranche book” at CIO was no “hidden prop desk”. If one looks at the picture of the first quarter of 2012, one may instead suspect that the bank records were overestimated then between Q4 2011 and Q1 2012 in quite a provocative way. The provocation leaves little doubt with the not so unfortunate “tempest in a teapot” statement. If any “mistake” was done then that was the SOLE responsibility of Dimon and members of the ALCO of JpMorgan. Was it a mistake in hindsight? The ‘london whale’ scandal actually brought back a better balance between many things including the real market price moves and the ALCO ultimate adjustments. As such the scandal acted as a “catharsis” (see April 17th call between Drew and Artajo for more information). But was it all, a “catharsis” at the end of the day?

It will be shown through the tangible equity analysis that the year 2012 was a unique bonanza year as a $25 billion generation of “tangible capital” arrived while the bank shrank its size. This sparked a very long awaited increase in the traded share price strongly supported by an exceptional increase in the book value that itself was plain “tangible”. This may have been much, much more than a mere “catharsis” for the bank….

The valuation process and the stakes behind it were described in the first part. Thus before the figures are lined up in the broader history of Jp Morgan itself, it matters to show where the reference numbers can be found in either 10-Q or annual reports. The figures displayed in the following tables were extracted in part from the annual report of 2011 that was disclosed on the 29th February 2012. One should switch to page 305 of this report and pick some of the numbers that will be used in the table displayed below. On should also turn to page 110 of the same report where the bank explains summarily what it means behind the labels “cash and due from banks”, “deposits with banks” and “fed funds sold and securities purchased under resale agreements”. The extracts below provide the descriptions:

“Cash and due from banks and deposits with banks---The Firm uses these instruments as part of its liquidity management activities. Cash and due from banks and deposits with banks increased significantly, reflecting the placement of funds with various central banks, including Federal Reserve Banks; the increase in these funds predominantly resulted from the overall growth in wholesale client deposits. For additional information, see the deposits discussion below. Federal funds sold and securities purchased under resale agreements; and securities borrowed---The Firm uses these instruments to support its client-driven market-making and risk management activities and to manage its cash positions. In particular, securities purchased under resale agreements and securities borrowed are used to provide funding or liquidity to clients through short-term purchases and borrowings of their securities by the Firm. Securities purchased under resale agreements and securities borrowed increased, predominantly in Corporate due to higher excess cash positions at year end. »

Then one should turn to page 179 of this annual report of 2011 in order to access some interesting data about the intangibles, but also about the exposures that were “eliminated in consolidation”. And on page 181 of the same annual report one can see the statement of cash-flows table that does NOT show clearly however that a massive inflow of cash totaling $30 billion occurred in Q3 2011. One has here to put all the figures into a spreadsheet to

35 see that event. This is not an event that will be detailed in this part although it is unique in JpMorgan’s history. But it will be mentioned as one among the precursory signs of a massive regime shift at Jp Morgan that reached its fastest realization pace in the first half of 2012 through the ‘London whale’ scandal actually.

On page 132 this time of the annual report of 2012, the bank states: “Cash flows from investing activities The Firm’s investing activities predominantly include loans originated to be held for investment, the AFS securities portfolio and other short-term interest-earning assets. For the year ended December 31, 2012, net cash of $119.8 billion was used in investing activities. This resulted from an increase in securities purchased under resale agreements due to deployment of the Firm’s excess cash by Treasury; higher deposits with banks reflecting placements of the Firm’s excess cash with various central banks, primarily Federal Reserve Banks;».

Pages 168 to 173 of the Annual report of 2013, brings the complete picture with a delay of this increase done “with various central banks, primarily Federal Reserve banks” and the one of “deposits with banks” with acronyms like “LCR”, “HQLA” and “NSFR”. One has to bear in mind that the US central banking system is the historical main regulator of JpMorgan. The keyword could also have been “tripartite-repos” for the experts of the 2008 financial crisis… This increase in “deposits with banks” results from an initiative that started right between the demise of Bear Stearns and the demise of Lehman Brothers, ie in June 2008. One may be able to retrieve an article of the FT of early July 2008 describing the quite active and ongoing involvement of just all the regulators worldwide to rein in this opaque “tri-partite” repo business. The regulators were quite active already on that matter of “cash at hands” knowing then that banks in general did not have enough liquidity reserves and capital provisions against their CDS exposures. And CIO was on the radar screen already in June 2008 as the “strategic liquidity reserve of JpMorgan” was invested by CIO in market-traded assets. Ina Drew was involved in the closed doors meetings on behalf of the whole firm and CIO at Dimon’s request.

It is enough at this stage to take note of the items that are circled as they will enter into the buildup of the following charts that will show the bonanza gain in itself.

The big picture of 2012 for Jp Morgan throughout the “London whale” event: a bonanza year

With all those references being available now one can start looking at the actual magnitude of gains that the “London whale” scandal brought to the firm so coincidentally. It has been mentioned that the event itself brought up about $25 billion if not more of “tangible capital” for the vintage “2012”. This is a minimum actually…. It will be seen that this figure was not here by chance. It matched with other figures and made the share price of Jp Morgan really more attractive than ever. The financial markets will by the way recognize this gain by making the bank traded share price skyrocket as early as August 2012 and onwards…

Let’s start with the surface of things: the S&P, the share price, the earnings….This history dates back to 1998, there is not enough room on the screen, so the table will be split in two parts:

Please notice the unique red figure here for year end 2011. The line that matters is the last one: it displays, assuming that dividends were re-invested on the shares themselves, the relative performance of an investment in Jp Morgan quoted shares versus the S&P 500 ( no re-investment of S&P dividends). The end of the year 2011 looks pretty bad. This is when CIO made a unique early and hasty year end close as of December 15th 2011. This is also when the Federal Reserve will worry about the projective unwind costs for the “tranche book” of CIO specifically in relation of share buybacks. Jp Morgan was then plagued with a lot of ongoing litigations, many of them related to the financial crisis of 2008. It matters to notice that the “book value”, ie the price inferred from the “common shareholders equity” and the actual number of shares, was following a multi-year increase since 2009. This did not seem to help much here…How bad was it if history was a guide?

The following table will show the former history where other red figures show:

Here one can notice the earliest moments when the share price of Jp Morgan had underperformed the market was in 1999. But then one must notice that the quoted share price stood above $51 while the book value was only at $18. Somehow a benign underperformance of the traded share price versus the book value could well be expected….The other moment when the share of Jp Morgan underperformed the market (represented by the S&P500) was in 2001 and 2002, right when the firm was plagued with the scandal of ENRON and the “option credit lines” issue. Still then, the quoted share price would be trading well above the book value.

In 2011 the share underperformed the market and quoted below the book value….Thus the end of the 2011 looked like a very harsh questioning of the very business model of Jp Morgan beyond the scandals. Unlike 1999, the traded share price was not any more sky high versus the book value, far from it. Unlike 2001-2002 there was neither an economic crisis, a market fall and even lass a massive scandal like ENRON to plague JpMorgan so specifically. Indeed the share prices traded below the book value and the performance of “Jp Morgan shares” as an investment lagged the market rise which in itself was pretty low at 6.2%. Worse this negative 5% underperformance was the cumulative number dating back from 1998. A lot of “value” was being destroyed constantly here. There was no visible reason for that situation other than a failing business model. Thus the many mergers (Chase, BankOne) and the many windfall acquisitions (Bear stearns, Washington Mutual, SIGMA, AIG bespoke deals, Lehman assets, Fortis assets and so on) seemed to have created nothing else but a dinosaur deemed to disappear sooner rather than later. That was a grim picture. But the “London whale” scandal would save the day so “cathartically”.

The chart below summarizes the picture as it looked at the end of 2011….

ENRON Cross current

London Whale !

How well timed this “London whale” scandal looks on this chart. This may just be the well manufactured result of associations made in hindsight with a biased purpose. It matters to cling to that assumption as long as possible in what follows….To realize that this was NOT a fortuitous coincidence of events. Nothing had gone out of control from that standpoint, quite the opposite. History is a reliable guide here.

One observes a fall between 1999 and 2003. This was mostly the impact of ENRON and related scandals of the time hitting JpMorgan in particular. The market and the share of Jp Morgan fell in synch but the share price fell more. This could simply be explained by its former overvaluation versus the book value. One also notices the cross-current between the end of 2008 and the end of 2011. Here the world economy and markets recovered. But Jp Morgan shares fell strikingly below their book value (while usually the very opposite occurs as investors turn more optimistic). Finally one can see the remarkable coincidence in timing between the ‘London Whale’ genuine scandal and the turnaround of the share price versus its book value and versus the markets trend in general. How providential this “London whale” scandal looks like all of a sudden! How much of this “miracle” was just a coincidence?

The following chart will emphasize in a more detailed way that this timing of the “London whale” genuine scandal was crucial as faithful investors were observing 13 years of investment in Jp Morgan turn sour. The chart shows the cumulative performance since 1998….At the end of 2011, the long term investors were seeing more than a decade of “value” vanishing in ashes. But worse, the markets indicated that the “book value” was too high, ie that either the books and records of Jp Morgan were fundamentally wrong or that the very business of JpMorgan was dead soon. And the “London whale” saved JpMorgan as it seems, far from being a damaging “incident”… Really, how much of all this was a “mistake”? On can doubt it was a “mistake” at all. But so far one should stubbornly stick with the view that it was just a peculiar coincidence.

London Whale !

One must have noticed the impressive progression of the “book value” here year after year. This is regular and looks like a secular capital creation trend. Yet, one has to remember that on the path the label “Jp Morgan” gobbled Chase in 2000, BankOne in 2004, Bear Stearns-Washington Mutual in 2008 along with a couple of other acquisitions. It will be analyzed later on this website the question as to whether this reflected a real “value creation” or actually some “value destruction” overall anyway. For example once one adjusts for the growth of the economy over the period and for the financial leverage risk that existed on Jp Morgan balance sheet, this book value increase is no so impressive. How much value was actually created for the USA or for the world economy here? How to answer simply the question? One first clue relates to the ‘tangible equity’ again or ‘tangible capital’ that has been created since 1998. This is what the following tables and charts will display.

As pointed out the brand “JpMorgan” actually supports a huge bank that has been built after aggregation of many former companies. It is complex and therefore “time consuming” to manage anyway. It matters to look first at the basic cost structure of this company that is not necessarily due to trading activities.

This cost structure includes:

* the “tax deferred expense/(Benefit)”: the firm regularly posts “net income tax amounts” every quarter. Yet the firm also draws either further reserves in that regard which is called a “tax deferred expense”. At times the bank may instead frontload some “tax deferred credits” which induces more cash inflows based upon a labeled “tax deferred benefit”. This is a provision or a release that is due to the firm complexity and size.

* the estimated effect of the non-consolidated parts in earnings. It is quite time consuming for the ALCO to make proper assessments. And it is quite time consuming too for all the watchdogs to thoroughly review these quarterly assessments of the ALCO members. The stakes are big for the bank, for the watchdogs, for the USA, for the world economy. People often sum it up through the expression “too big to fail”…. The bank has huge exposures indeed that are NOT consolidated or are “eliminated” in the consolidation process. Why is that? Typically the reason, as stated by the bank, is that it does not fully control the valuation and evolution of those “off-shored” exposures. In practice this provides some randomness on the actual performance of the bank, a thing which the ALCO adjusts every quarter in a non transparent process as well. See ‘other receivables’ and ‘other collateral’ here in the “wholesale credit portfolio”….

* the fundamental “estimated fair value” uncertainty that is monitored through the latest ALCO stage: as explained, the ALCO sets the ultimate ‘fair value’ including liquidity provisions that are NOT disclosed as such for example with regards to the breakdown between “level 3” and “level 2” especially for credit derivatives. As the firm states, this process is NOT transparent (most likely for strategic reasons) but it is related to “basis risk” no doubt.

* The ultimate impact that all those costs or random effects have on the balance sheet is easy to see: the asset value and the liability value do not move in tandem from one quarter to the next. Thus, once the ALCO has first adjusted the derivatives fair value as such, it also has to adjust the total asset ‘carrying value’ to the total ‘liabilities carrying value’ so that they match. The ALCO can do that in multiple ways. Watchdogs and board members do watch this out!

In what follows, there will be technical comments based upon tables. This is just a preliminary stage before showing the gain in “tangible capital”. These comments are meant to show the magnitude of the changes that occurred through 2012. It is meant to show as well how these high structural cost would be translated in the books and records of the firm while the “London whale” scandal was unfolding. This is not as such the proof for the bonanza gains. It is just ultimately meant to show that the plan of Dimon consisted altogether in managing these “too big to fail” structural costs, ‘winding down” the “tranche book of CIO”, and indeed creating a massive tangible capital inflow for the firm. And here we start with what Dimon wanted to address and how these key figures popped up in the ledgers of Jp Morgan.

The table below provides some estimation of those structural factors based on the information made available by the bank in its annual reports (10-K) and quarterly reports (10-Q). The history is incomplete as only the most recent regulatory requirements allowed to have just a more complete picture. When needed the numbers are specified to be in billions, as they show on the reports themselves. The “tax deferred expense/(benefit)” was available prior to 2004, but it has only some relevance since Jamie Dimon took over in the light of the ‘London whale’ July 2012 restatement (see prior documents posted on this website through “for those who wonder…”).

The table above shows interesting features especially the results of the SFAS 107 reporting rule on the last line of the table. One should not try to draw conclusions yet here but just observe some significant changes as the years passed. They will see that there are some gains, some additional big provisions and some conservative adjustments that all lined up AFTER the “London whale” scandal had unfolded. As such they suggest a gain that came on top of the earnings that would be reported to the public. All this occurred across the years 2012- 2013 and 2014. It is not yet to say that they all related to the “London whale” scandal. Yet….

As one will see, the upturn in the former chart about the share price is corroborated here in this table by gains of $20 to $30 billion. There gains that are observable but they are spread across different spot of the bank’s ledgers. One has to bear in mind that the year on year record profits as reported after the “ALCO STAGE” ranged between $18 billion and $ 22 billion between 2010 and 2014:

* for the “tax deferred” amounts: the firm actually took unusually high cash amounts as a provision in 2013 and 2014, right after the ‘london whale’. These amounts were as much money that was NOT reported in earnings since they were straight deductions from the gross revenue. These amounts came on top of the current ‘net income tax’ and were NOT litigation provisions. One can notice that the opposite occurred in 2008 and 2009. Thus those figures suggest that the senior management expectation was of “lower profits” for 2009 and 2010, and the expectation swung radically in the opposite way in late 2011. If one sums the provisions on tax from 2011 to 2014, one gets to about $21 billion, or a full year of record profits that has been withdrawn over 4 years here. The spike in provisions occurred in 2013 while the bank was harshly criticized, scrutinized, forced to divest, fined heavily…but still in the anticipation of further higher profits looking forward as common sense suggests….

* If one searches for the ‘other cash and other receivables and other collateral’ exposures that the firm has but not in the “legally enforceable” context of customary master netting agreements, one can also observe big swings, especially in 2010 ($16 billion), and in 2012-2013-2014 (resp $9.99 bln, $12.31 bln and $9.37 bln). These are pretty material swings of value for a firm that produces about $20 billion profits per year at the time….As much as the gain of 2010 could be attributed to the markets’ overall recovery that induced the reduction in cost of the basis risk and skew risk, the following gains for 2012- 2013-2014 look very much the same. The total gain in this section for the 3 years that are linked to the ‘london whale’ scandal here reaches about $31.5 bln, or about 1.5 year of record earnings. These are massive gains overwhelming the $6 bilion loss at CIO, the $5 billion profit printed for Q2 2012…

*So one may sum up saying: “ well there are $21 billion of tax provisions taken ahead of time AND there is $31 billion of gains potentially unreported yet… That could total $52 billion right?”. Not quite. Who knows actually? May be the ALCO stage was overly optimistic in the backyard…. If one switches now to the “level 3” related ultimate ALCO-driven adjustment that is done on the derivatives fair value, one first should notice the magnitude of the swing that reversed itself between 2008 and 2009. The ALCO turned very conservative on derivatives in 2008 and swung back to a very aggressive stance in 2009. Where was the “money” that justified such a swing in the senior management opinion at the ALCO stage? No doubt this sparked the very acute attention to regulators in both cases. But overall the impact was nil almost for the combined 2008-2009. What about the period covering the ‘London Whale’, ie 2011 till 2014? The Net impact of the ALCO is a negative $3 billion in the reported carrying value change over the 4 years. Pretty flat to conservative… Thus, although the ALCO looked relatively aggressive on its Level 3 assessment for 2012 with $8.8 billion overestimation, but this was balanced with 2011 and 2013. Therefore on may conclude by saying that the provisions or the “non consolidated” gains mentioned above came along with a rather conservative impact of ALCO for derivatives as such.

*One may remember the SFAS 107 rule then… Once all those factors have been included in the balance sheet computation and the earnings production, what was the ultimate adjustment that the ALCO set in both assets and liabilities, aside from the derivatives themselves? The last line on the table above displays the details. Here one cannot sum up the adjustments over the years since they are made relative to the computed output year after year. Therefore the adjustment of the year before is reset to zero before the ALCO repeats its own final exercise. It matters to notice in 2008 that the ALCO overestimated the assets by $21 billion and the overall earnings by $15.5 billion from what the firm- wide computers showed. Also the ALCO that year undervalued the derivatives by $23 billion but overestimated the other “cash” assets by some $21 billion overall. In 2009, the trouble was that the ALCO overestimated the derivatives value by $22 billion AND overestimated its earnings by some $7 billion overall. Was the “principal transaction” result correct? No wonder the regulators looked in depth at the VaR analysis of Jp Morgan that year. A very similar pattern occurred in 2012 actually where the earnings looked overinflated by some $5 billion. The overestimation went even higher for 2013. And the regulators then were watching Jp Morgan more than ever if only because of the ‘london whale’ scandal. And they approved those accounts while blaming the bank all the way still. But they also certainly scrutinized the “non consolidated” results before approving such over-estimation on paper. That earnings production was conservative anyway in the regulators eyes despite the massive headline $13 billion fine that the bank paid in late 2013. What is the difference between 2009 and 2012? The ‘tax deferred’ figures and the ‘other cash other receivables other collateral’ give a clue. The firm is harvesting both $21 billion of tax related provisions and some $31 billion of yet unreported gains somewhere out of the ‘legally enforceable’ procedures. The bank seemed to walk a tight rope right? Things would settle nicely in 2014 where the ALCO resumed a standard conservative stance and the bank kept printing sky high profits. This alone shows that regulators saw those gains coming right when the earnings were already setting new records. They missed none of the ALCO adjustments. They supported the way the bank would report its figures although the headline statements did not convey the root causes for the upturn…And this root cause dated back to the “London whale” scandal.

To make this last picture clear, the following table sums up what both the ALCO, the board and the regulators saw then. The first lines just repeat the former table for reference purpose with additional few more lines. The added last 4 lines give the reported earnings each year, the earnings as the ALCO grossly saw them before doing its final adjustment on the balance sheet. The line below provides an estimate of what the ‘other’ derivatives exposures performance was on the year (this is based on the ‘other collateral’ figures usually available on the “

45 wholesale credit portfolio” tables in the 10-Q and 10-K reports). The last line shows what the “gross” estimated earnings figure was before the ALCO adjusted derivatives and balance sheet gross figures. Such a calculation could easily be performed by any watchdog watching. And they were watching since late 2011:

This $52 billion total of “unreported gains” that the ALCO would adjust rather conservatively on the whole year after year between 2011 and 2012 is not conclusive though. This is just a series of observations as such. But it shows figures that were also well “noticed” by all regulators. They arose in the books and records around the “cost structure” of this “too big to fail” bank. They could not be missed by the regulators anyway. And they would be scrutinized if only because they had served as flashing red lights through the last financial crisis.

One should look at the numbers for 2008. The picture was not so rosy then (see “- 9 895” as circled above). Thus when Jamie Dimon claimed to repay the TARP money in Q2 2009, regulators were NOT comfortable at all. By the way, the 10-Q reports show that Dimon paid back the TARP money then by issuing shares and long term debt. Thus, Dimon did NOT have extra cash to pay back the TARP money in 2009. One can see that the year 2010 was instead very profitable on paper, just as if a miracle had occurred…. Then Dimon launched the share buyback program which should have turned well. But in 2011, the skew and the basis risks costs increased and the unadjusted earnings for the firm looked close to nil actually (see now the “384” figure 2 lines down as circled above). This is when the ‘London whale’ scandal burgeoned in the markets, as early as January 2012 as the public reports say today. And by an act of providence, the end of 2012 unadjusted earnings would be sky- rocketing at $30 billion. The storyboard of the “ALCO stage” can be sketched albeit speculatively so…. There was some reserve taken out of the unreported gains here in 2012….No wonder, the ALCO could afford to be a bit aggressive (unlike 2009) and the bank could take provisions and the ALCO could also overestimate the ‘carrying value’ of derivatives…. And the regulators could let Dimon do all this in 2012, in 2013. In 2014 all

46 would settle in with $21 billion of provisions made –maybe or maybe not- on a pending $31 billion gain that showed just with the ‘other cash, other receivables, other collateral’ managed through contracts that were not part of ‘legally enforceable master netting agreements’. Thus with those 4 lines one can see very clearly how 2012 differed from 2009…

One can see here that the upturn through the “London Whale” for the traded share price was not so much a coincidental event. Back in 2009 the regulators were very worried for good cause as Jp Morgan lacked extra cash despite its massive CIO and its massive “synthetic correlation” hedge. In 2012, the regulators were very relieved for good cause seeing these $31 billion of pending gains coming at last, Jp Morgan being flushed with $30 billion extra cash reserves, and $21 billion of provisions for tax related purpose. Why would Dimon make these $21 billion tax related provisions if he was not ALREADY anticipating much higher profits than the ones officially printed? How conveniently and quietly these $21 billion of provisions were used to allege that the bank had been hammered by these unfair litigations and this allegedly unwanted “London whale” scandal.

The regulators in 2012 and 2013 definitely were watching all this evolution in the middle of what Preet Bhararat (DOJ) would characterize as a “perfect storm of misconduct” pervading throughout all the ranks in the hierarchy. This must mean that a “perfect storm of scrupulous scrutiny” pervaded throughout all the ranks of just all the regulators as a response. The authorities must have had this time some quite “tangible” reassurances from the bank. This is what has just been shown above. The regulators have remained mute all along on this already quite “tangible” certainty here that Jp Morgan projected much higher profits right after the “London whale” event. Instead they have all endorsed the thesis deployed by Dimon that the top management had been surprised and not informed well enough, escaping a big loss by a hair. One may understand many “motives” for the bank to do so, either honest or less honest. Still it is really hard to imagine why all the regulators endorsed the bank’s version on the “calamitous trading loss” line.

There is still a possibility that the “tangible massive gains” were happening in parallel to the scandal, independently so to some extent. Who heard such a message ever other than quite ambiguous reassurances that “the profitability of the bank had remained solid”? The bank would never try to disentangle its profits from the outcome of the “London whale” scandal. Nevertheless the figures displayed above are no sign whatsoever that the bank was harmed as such by the scandal in terms of profits and other quite ‘tangible’ gains. Opacity would prevail all along.

Thus in what follows the reader will just have one analysis based on the “tangible capital”. That will show a big gain that was created right then through the “London whale” scandal itself. One will see how this allowed the shareholders at last, after 12 years of patience, to feel that their investment had a strong potential to gain in value in the coming years.

The reasoning starts with the traded share price itself. Is it too high? What could make it go higher? This is the expected “reward” that shareholders should expect to get. But there is a risk, as always. Shareholders have to weigh “risk” and “reward” as much the bank executives did all day. What are the risk and the reward in $ price on the share price itself as it trades in the markets? The risk is that a recession occurs, a bad headline news pops on “Jp Morgan”, or a regulatory ruling goes against the bank. Then the market price would crash more or less, ie it would drop suddenly and lastingly to a much lower figure.

How much could it crash then? Well this bank is not worthless right? As explained formerly, one figure in the consolidated (but unaudited) balance sheet shows the “shareholders common equity”, which is equal to the “book value” of the shares times the number of shares in circulation. The “book value” is the first one tangible floor price that indicates “where” the market price should start stopping the fall….Unless the “book value” is

47 itself based on assets in the balance sheet that may also change radically in value should a bad news hit the bank. This would likely be the case of any “intangible asset” or some opaquely defined “asset”. And here one must assume that the “book value” contains almost all of the “intangible assets” and “other assets”, among which one primarily has the “goodwill”.

This is not a benign consideration in the case of Jp Morgan since the merger with BankOne (2004). Since then the ‘intangible assets’ amounted then to 55% ($57 bln vs $105 bln) of the common equity and shrank down only to 33% in 2011 ($58 bln to $175 bln). This remained quite a big share of “intangible” value. Shareholders have to accept that they shoulder the whole risk around the “intangible assets” of the bank. But up to what point should they? Really it is up to them right? Or actually is it also up to Dimon to revise down the value of his “intangible assets” guesstimates and to therefore revise down the “book value” as well by the same token? Or is it actually the duty of regulators to have their say as to the “value” that the CEO of JpMorgan has actually created over the years? If the scandal lying behind Bernard Madoff is a guide, the answers are crystal clear.

Still the investors in Jpmorgan shares need a floor for their own use and they need to construct a “reward” scheme for themselves. They would not know anyway what Dimon and regulators discuss behind closed doors on “intangible assets” or “other assets”. Thus the “tangible equity” is quite a key benchmark for the shareholders. They should skip this amount of “intangible assets” first and focus on what could trigger better projections for the future based on the tangible assets sitting on the balance sheet as described before. What should this “tangible equity” be compared to then? As explained before, one should wonder what kind of assets the other stakeholders of the bank would not want to rely on. The category “other assets” listed as such on the balance sheet is a good candidate as by definition it is not well identified, not clamored as being a great liquid holding, labeled as such by elimination of the other well branded asset classes for which there is a known market with regular reliable prices.

Thus equity investors should really consider that the “typical” assets serve to pay the debt, the depositors, the employees, the routine bills, the tax, the other authorities etc… And the “other assets” are the routine background for their “tangible equity” to be paid back one day if needed. If the “book value”, of which one has removed the “intangible assets”, is worth more than the opaque “other assets” class, then the shareholders can benefit from the well flagged growth and profits coming from the well defined assets classes listed on the balance sheet. But this projection is only possible if the “tangible equity”, being worth more on paper than the “other assets” category, starts being “justified” by well identified assets. Instead, as long as the “tangible equity” is worth less or slightly the same value than “other assets” than the share price, assuming that intangible assets remain stable, investors are in the blind. Betting that the traded share price would go higher then is highly speculative in itself for investors as they depend here on confidential meetings between the CEO and its watchdogs. In such case most likely the bank growth is not going to translate into higher share price expectations for the Jp Morgan traded shares.

If the tangible book value solely relies upon the “other assets”, investors may even feel scared and sell their shares at a discount to the “book value”. They would here simply fear that “intangibles” are just that, intangible that can vanish in a second. Thus the difference “ ’tangible equity’ – ‘other assets’ “ is quite a good trigger to project a strong rise of the actual traded share price. It may as well induce a traded share price to collapse below the book value lastingly. At the end of the day, one sees here how much the “personality of Dimon” matters in making credible the “intangible assets” value. One has to bear in mind that if the actual traded share price stands lastingly below the “book value” then market players and shareholders might wonder first whether Dimon did estimate properly the “intangible assets” including his own business strategy. Next they might wonder whether Dimon did estimate properly the “opaque and illiquid” “other assets”. For want of clear answer from Dimon, they may well anticipate a bankruptcy or at least a collapse in the “intangible value” which amounts to the same

48 final outcome in the case of Jp Morgan since 2004. Indeed for Jp Morgan since the merger with BankOne, the collapse of the “intangible value” most likely would mean a bankruptcy soon to come. Thus the stakes behind this difference “ ’tangible equity’ – ‘other assets’ “ were vital to Dimon, vital to the regulators and crucial to the shareholders of JpMorgan.

What were these “other assets” in the very language of JpMorgan? It is first useful to quickly check what the bank Jp Morgan itself states about the “other assets” class that is displayed on its “consolidated balance sheets (unaudited)”. If one runs a search in “other assets” on the May 10th 2012 10-Q for Q1 2012, one will first see that the bank refers here to either “private equity investments”, or “disputed assets” or “predominantly margin loans” (when they contribute to the net interest income). On page 173 one can read the following: “CB revenue: Lending includes a variety of financing alternatives, which are primarily provided on a basis secured by receivables, inventory, equipment, real estate or other assets. Products include term loans, revolving lines of credit, bridge financing, asset-based structures, leases, commercial card products and standby letters of credit.”

It is worth looking at the annual report of 2011 next on page 110 to see how the bank listed its different asset classes so that one figures that the “other asset” label basically gathers all the sources of revenue that do not perform as typical sources of income (see the table after the current comment). On the following page, the bank describes what the label “other assets” contains: “Other assets Other assets consist of private equity and other instruments, cash collateral pledged, corporate- and bank-owned life insurance policies, assets acquired in loan satisfactions (including real estate owned), and all other assets. Other assets remained relatively flat in 2011.” On page 136 next of the same 2011-Annual Report, one can find a clue to what those “other assets” are intuitively speaking in general.

There, on page 136, the bank discloses its overall exposure to credit markets via a “wholesale credit portfolio” label. This is where the bank discloses additional “receivables” and additional “collateral” that are NOT counted within the standard fair value determination process for credit derivatives in particular. These amounts are NOT part of “legally enforceable” master netting agreements. They do have an impact on the ALCO ultimate adjustments be they either linked to the “level 3” adjustment or be they linked to the final balance sheet adjustment (reported under SFAS 107 rule). For the additional receivables, the bank labels them as “Receivables from customers and interests in purchased receivables(a)”. and the reference to the footnote “(a”) states that, “(a) Receivables from customers primarily represent margin loans to prime and retail brokerage customers, which are included in accrued interest and accounts receivable on the Consolidated Balance Sheets. “Interests in purchased receivables” represents ownership interests in cash flows of a pool of receivables transferred by third-party sellers into bankruptcy-remote entities, generally trusts, which are included in other assets on the Consolidated Balance Sheets. » On page 182 of this 2011 annual report, the firm explains how it consolidates the “VIE” (VIE for Variable Interest Entities, that all sit out of the balance sheet of Jp Morgan by original design- the firm specifies that many a re-consolidated-) assets in “other assets”: “Investments in companies in which the Firm has significant influence over operating and financing decisions (but does not own a majority of the voting equity interests) are accounted for (i) in accordance with the equity method of accounting (which requires the Firm to recognize its proportionate share of the entity’s net earnings), or (ii) at fair value if the fair value option was elected at the inception of the Firm’s investment. These investments are generally included in other assets, with income or loss included in other income. »

On page 234, Jp Morgan explains that when it takes actual possession of a foreclosed property, it is recorded then in “other assets”. On page 274, Jp Morgan further states :” Junior subordinated deferrable interest debentures held by trusts that issued guaranteed capital debt securities At December 31, 2011, the Firm had established 26 wholly-owned Delaware statutory business trusts (“issuer trusts”) that had issued guaranteed capital debt securities. The junior subordinated deferrable interest debentures issued by the Firm to the issuer trusts, totaling $20.9 billion and $20.3 billion at December 31, 2011 and 2010, respectively, were reflected in the Firm’s Consolidated Balance Sheets in long-term debt, and in the table on the preceding page under the caption “Junior subordinated debt” (i.e., trust preferred capital debt securities). The Firm also records the common capital securities issued by the issuer trusts in other assets in its Consolidated Balance Sheets at December 31, 2011 and 2010. »

The annual report of 2012 will not provide more information about what “other assets” means. But the 2008 annual report explained that many of the derivatives margin loans of Bear Stearns and Washington Mutual ABS related deals had been moved to “other assets”. These were all basis and skew risks as the bank discloses it with its own words on page 77 of the 2008 annual report: “Other assets The Firm’s other assets consist of private equity and other investments, collateral received, corporate and bank-owned life insurance policies, premises and equipment, assets acquired in loan satisfaction (including real estate owned), and all other assets. The increase in other assets from December 31, 2007, was due to the Bear Stearns merger, which partly resulted in a higher volume of collateral received from customers, the Washington Mutual transaction, and the purchase of asset-backed commercial paper from money market mutual funds in connection with the Federal Reserve’s Asset-Backed Commercial Paper Money Market Mutual Fund Liquidity Facility (“AML Facility”), which was established by the Federal Reserve on September 19, 2008, as a temporary lending facility to provide liquidity to eligible U.S. money market mutual funds. For additional information regarding the AML Facility, see Executive Overview and Note 21 on pages 41–44 and 202 respectively, of this Annual Report.”

One can grasp a bit more intuition actually through the 2013 annual report on page 46 here : ” Other assets The increase is primarily driven by the implementation of gross initial margin requirements for certain U.S. counterparties for exchange-traded derivatives (“ETD”), higher ETD margin balances, and mandatory clearing for certain over-the-counter derivative contracts in the U.S. “ This comment of the bank here indicates that although the “other assets” conveyed some margin loans, basis risks, skew exposures, the bank did NOT request a full clearing process. This suggests that before 2013, the bank could hold gains through the “other assets” for related derivative exposures that went unreported at the time simply because there was no “clearing” process in place similar to the one that CIO had say through ICE or through the centralized IB collateral management. On the same 2013 annual report, the bank makes a special mention to transferred exposures that were NOT “qualifying for sale accounting” (page 227): “Transfers not qualifying for sale accounting In addition, at December 31, 2013 and 2012, the Firm held $14.6 billion and $9.6 billion, respectively, of financial assets for which the rights have been transferred to third parties; however, the transfers did not qualify as a sale in accordance with U.S. GAAP. These transfers have been recognized as collateralized financing transactions. The transferred assets are recorded in trading assets, other assets and loans, and the corresponding liabilities are recorded in other borrowed funds, accounts payable and other liabilities, and long-term debt, on the Consolidated Balance Sheets. »

It is not to say that this refers straight to the transfer and collapse of synthetic tranche positions that Dimon launched in 2010 and finalized through the “London Whale” scandal. Because, had it been the case, the bank would certainly have specified it, right? But this comment here further explains what “other assets” may contain and how close they look “business wise”. This describes a bit better what the “tranche book” at CIO was expected to hedge day after day for the bank. One can see that if the “tranche book” of CIO, sitting on the consolidated balance sheet, and being subject to immediate “legally enforceable” contracts was naturally mostly hedging “de-consolidated” or “eliminated in consolidation” or “not legally enforceable” or “transfers not qualifying for sale accounting” risks”, the very performance of this “tranche book” of CIO could only be nebulous, overridden all the time and actually unrelated to the actual “estimate P&L” report that the London CIO staff was anyway mandated to process as the firm wanted it.

It is time to get back to page 110 of the 2011 annual report, and “see” how those “other assets” sit in the balance sheet reports…..

Now that the profile of “other assets” is clarified, it makes more sense to observe their history in the balance sheet and see how they matched against the “tangible equity” over time across the “fortress balance sheet”.

The table below will show now the history of those “other assets” in relation to the earnings printed by the firm.

Period Q4 2014 Q4 2013 Q4 2012 Q4 2011 Q4 2010 Q4 2009 Q4 2008 Q4 2007 Q4 2006 Q4 2005 Q4 2004 Q4 2003 Q4 2002 Q4 2001 Q4 2000 Q4 1999 other assets in Bln 102 950 110 011 101 775 104 131 105 291 107 091 111 200 74 314 51 765 48 195 45 600 52 573 44 814 50 205 24 530 20 440 Net interest income on other assets in Mln 663 538 259 606 541 479 895 652 933 291 64 margin in Bps 0,64% 0,49% 0,25% 0,58% 0,51% 0,45% 0,80% 0,88% 1,80% 0,60% 0,14%

One can see here that as Jp Morgan made more acquisitions and mergers since the end of 1999, the “other assets” category ballooned from $20 billion to $110 bln in late 2008. This testifies of a massive increase in “basis risks”. It was a welcome increase to avoid raising further the value of “intangible assets” to justify the steady rise of the “book value” over time, wasn’t it? In the same period of time the “common shareholder equity” increased from $35 billion in late 1999 to $135 billion in late 2008. Thus while the common shareholders’ equity increased by 400%, the “other assets” increased by 550%. The picture is even more worrisome when one removes the intangible assets. Indeed the “tangible equity” grew from $25 billion in late 1999 to “only” $72 billion at the end of 2008, which amounts to a meager 290% (versus 550% increase in “other assets” as pictured above). Many merger and acquisition would fuel the external growth of the brand JpMorgan over the period. These many “golden deals” looked as good justification for the “book value” to rise. But how much should it rise? How much “tangible value at the end of the day” was being created through those massive tectonic shifts of capital?

One can see that a major increase occurred in the course of 2008 for the “other assets” as a result of that. And when one looks at the most profitable period for those “other assets”, it was in 2006 and 2007 actually. This

51 margin spike coincides pretty well with the boom and bust of the ABS-CDO-Tranche-CDS-Subprime markets. This spike coincides very well also with the history of credit derivatives. Back then banks piled up “basis risks “ and skew risks” one over the other and compressed them to the maximum. The notional amounts followed then an exponential growth and were already reaching more than $15 trillion….Yet the skew risks and basis risks were not at zero. They would not be close to zero despite the seeming bust of liquidity that came with the genuine explosion of notional amounts. It was just an illusion of liquidity that fooled no one among regulators and major CDS dealers. When the exponential growth of notional amounts in CDS would stop, the illusion would show for what it was…. And it would end badly… That was in 2006. This is also when the bank started ramping up this “tranche book” at CIO searching a protection against an explosion of the “basis risks” and the “skew risks” altogether…. This is just another confirmation of the relation that existed for years between this the book value, “tangible capital”, “other assets”, the “basis risk” on CDS and the “tranche book” of CIO.

It is now useful to compare the “ coincidental timing” of this increase in “other assets” and the associated margin. The table below will thus complete the former one above:

One can see that in 2004 the goodwill explodes, and the intangibles explode. And in 2005 the “other assets” margin explodes but not the notional amount of “other assets”. The bank then structured, not CDO, but CDO- squares, CDO-cubes, SIVs: it was piling basis risks and manufactured financial leverage on them. Where would these “structuring assets” be located in the balance sheet? Who knows….This suggests that riskier “basis risk” was being warehoused in “other assets” on purpose at JpMorgan-Chase-BankOne. Bear Stearns, Lehman, WAMU and others were still alive and well then. These are the very same “CDO-square” and CDO-cube or else structures that would cause the demise of the financial institutions through the crisis. These are the assets that will gross up the size of “other assets” at Jp Morgan starting in 2008 as the “golden deals” piled up.

Was the bank targeting the future golden asset class or was it showing its presence in order to avoid a crash already in 2006-2007? How profitable those investments in such “other assets” were? What is striking in that regard is to see the “impact” in net interest income terms: the impact jumps from almost zero to almost $1 billion between 2004 and 2006. This is as such a big increase but it does not weigh much in the balance of the big bank. Here is a table that the bank disclosed in its annual report of 2006 on page 28 provides more information on the matter:

Now one can see that in the course of 2006, the “Net Interest Income” (or NII) impact of “other assets” grew by $640 million while the total NII grew only by $1 687 million. One can also notice that the Non-Interest revenue grew even faster than the NII. “Net Interest Income” was insufficient anyway when compared to “Non Interest Income”. Since the merger with BankOne there seemed to be a strategy around “other assets”, that regulators saw developing, and it was NOT working so well. This strategy can be summed up as follows: there was a huge creation of “intangible value” that was to be amortized through the generation of profits using riskier and larger “other assets” at the new banking group “JpMorgan”. It was precluded then to make depositors or other debt- holders pay for the amortization of the intangible assets. The idea most likely was to increase the market share on structuring CDOs using the “goodwill”, all this risk being borne by shareholders.

But in 2006, it turns out already that the plan will NOT work. The operating costs of the behemoth and its size were too big already.

That was a structural problem at Jp Morgan since 1998 actually and the merger with Chase and BankOne did not seem to have solved the issue here whereby Jp Morgan depended more and more on market liquidity and unwind costs rather than predictable streams of income. It definitely was more and more too big too fail. Here the “tranche book” of CIO would be born as projected in 2006 coincidentally so. And it was not helpful at all in that it was only increasing the “unwind costs” and was NOT adding to the NII itself. It was known in 2007 by the firm and the regulators, ie from the very start. The bank could not exit. The bank had to wait. The bank sat on too much intangible assets and insufficient Net Interest Income. How to prevent the next crisis from pushing JpMorgan in bankruptcy? For one “Build a visible defense line” through a portfolio like the “SCP”. It would not be in the IB. It would be “top secret”. It was not to stay longer than needed…And for 2 use CIO to harvest as much “Net Interest Income” as possible. That may have to stay longer than needed….Regulators blessed that move…But they would not hunt to the “post implementation review” of the NBIA of 2006 that would preside over the birth of their “SCP”…

One could suggest at this point that “supposedly” the “SCP” project in 2006 was already dead-born for the watchdogs and Dimon alike: they would just be pretending to place a patch, visibly so for any market player to see, that was meant to protect the recent frenzy around the “other assets” that had started in 2004 with the arrival of BankOne. This scenario would be well corroborated if in the regulators files one found that the coming crisis in CDS was heavily debated behind closed doors with Dimon in person no later than 2005. It was a debate on the lines of “JpMorgan had to do something about it, preventively and above all visibly for all the market players to see”. Of course it could not be “the regulators’ idea”…. And here it was all about targeting trades that actually other market players would need for sure if they had themselves to unwind in emergency situations. It sounds shocking doesn’t it? It was a big “mistake”. This would then “supposedly” lead to think that regulators were quite active designers of the “tranche book” of CIO but really secretly. Officially of course, this “tranche book” of CIO would NOT be a “dedicated hedge” as such. It would NOT have clearly stated assets to protect for. Thus the bank and the regulators would NOT have had to disclose too much of this initiative. Logically so the NBIA post implementation review would NOT have to be even started if only to avoid leaving footprints. No this “initiative” would be a “book” that would have no name, that would be placed under the shelter of a “no risk” entity like CIO and would vaguely be meant to address “structural risks of the balance sheet” rightly so….until it was to be “taken down”….This “book” would therefore have a birth certificate that would be left unstamped in the civil registry, this being in plain violation of the Sarbanes-Oxley laws of 2003….This is just a working assumption, but one that is easy to verify in the files of 2005, 2006 and 2007.

From the different references disclosed by the bank about what “Other assets” were, one may assume that they contributed even more to the “Non Interest Income” bucket than they did on NII actually (to be sure “NII” stands for “Net Interest Income”). These figures and tables indicate that there had been here a definite strategic choice initially to grow the margin on “other assets” and their contribution to “Net income” be that through “NII” or through Non-Interest income.

Thus, knowing now what “other assets” are and what the firm’s strategy towards them was since 2004, it is useful now to look at the evolution of the “tangible equity”, the rock-bottom floor in the book value that shareholders might care about, versus the evolution of those “other assets”. This is what the following table displays:

One should start noticing the very significant increase in “common stockholder’s equity” between 2003 and 2004 that is of about $50 billion while the intangibles increased on the same period by $43 billion. The “other assets” had decreased then by $7 billion. Looking a bit further down, one can see that the merger with BankOne, ended a 3 year period during which “other assets” were worth more than the tangible equity. So those figures indicate that there might have been an arbitrage here that was done between the massive increase in “intangible assets” and the fact that the resulting “tangible equity” was now a bit larger than the “other assets” value thanks to the arrival of BankOne. That indeed left room to increase the riskiness of “other assets” and their size altogether. Was it sound? That was in 2006 and as shown above, the next move of Dimon would be to grow the

“other assets” revenue line even further while hurriedly deploying the “tranche book” of CIO for the next year imperatively.

Now one should look at the years 2007 till 2011. “Other assets” totals moved up from $46 billion to $105-110 billion. The contribution to NII had declined despite the entry “for a dime” of Bear Stearns and WAMU assets as the annual report of 2008 mentioned. Was there a provisioning of some sort that went undisclosed in the 10-Q and 10-K reports. Were regulators blind here? In 2010 the economy and markets had recovered but the book value of the share, cleared of its “intangible assets” component remained below the value of those “other assets” that did not perform so well on a predictable stream of income basis (NII).

It did not feel like the “fortress balance sheet” was stronger. It actually felt the very opposite. And regulators made their concerns quite official right then in late 2010. Were they really discovering anything new here? Coincidentally the traded share price settled below the book value. Was it so surprising? The year 2011 closed apparently with a strong improvement of the “tangible book value” relative to “other assets”. It looks welcome then, doesn’t it? But as mentioned before, the evolution of “other receivables” and ‘other collateral’ that are NOT fully consolidated ( or eliminated in consolidation---see the “wholesale credit portfolio” figures) did not give such a reassuring picture as the following table reminds. The former tables showed that the unadjusted earnings for the whole year 2011 looked rather like “zero” than anything…..

As mentioned, “other assets” contained some of the economic performance of some of those positions that were sitting outside of the usual “consolidated balance sheets (unaudited)”. Thus the welcome improvement on paper did not sound so real. It must be another coincidence if in early December 2011, CIO and the IB differed in skew prices that led to a $300 million difference about the “tranche book”. It is no doubt pure hazard if the IB traders acknowledged that they did not mark their tranche position where CIO saw the “bid-offer” prices on its end. It must be just one more coincidence if the CIO ran a first early year end valuation for the “tranche book” as of 15th December. It is mind you also a coincidence if, after the VCG price control group had validated CIO prices and communicated the result to regulators by the 21st December 2011, the Federal reserve sent a request about the “unwind costs” on this “tranche book” on the 22nd of December 2011. And it is pure randomness in this complex bank if the CIO ran another year end valuation as of 31st December 2011.

In the meantime it was reported to the higher ranks of the firm that unwinding the first 25% of the “tranche book” ($10 billion of RWA over a total $43 billion measured RWA the time) would cost CIO about $1.1 billion or more maybe in 2012. Thus the official picture at the end of 2011 looked good but it was neither convincing inside the bank nor outside the walls of the fortress as for the Federal Reserve for example. As the US Senate report described, the OCC examiners had been told that the “tranche book” would be taken down. So the OCC cared about the imminent fate of the “tranche book” of CIO. So did the FCA which would never complain that it was unaware that the ‘tranche book’ of CIO was to be taken down like “soon”. As per the dictionary definition of “to take down”, this meant that the “tranche book” was to be dismantled. This alone called for a $4 billion liquidity reserve as a “drawdown” reserve, irrespective of “how” the book would be “taken down” in 2012. Where was it taken in 2011? The table that follows reminds the figure that have just been mentioned here.

Fortunately for the bank Jp Morgan, through the “London whale” scandal, the “tranche book of CIO would be dismantled indeed along with the whole CIO as a structure. And this time at last the “tangible equity” will largely be worth more than “other assets” as the next table shows:

One can see that this will be because the tangible capital increased abruptly right in 2012 and further into 2014. One will also see that “intangibles” will be amortized on the way but not a lot. Thus a genuine “tangible value” had been created through this “London Whale” scandal. It was recognized to at last allow the “book value” to grow while amortizing the “intangible assets” value. Who does that so easily in normal times? Just nobody, not even Dimon as his own track record shows…

The next table will show that the firm will return quickly then to more acceptable ratios in broader terms of “equity versus assets” and that it will be able to release provisions for credit losses. This is another tangible sign that the firm was deleveraging its balance sheet in a truly reassuring fashion for the watchdogs. Investors could only sheer up then. Therefore this strong increase in tangible capital was synchronized with other very positive signs, right in 2012, despite this “London Whale” scandal….or thanks to this scandal

Compare for example the 6.2% “equity/assets” ratio and $23 Bln provision for credit losses in 2008, versus the 8.27% ratio and $16 billion of 2012. All the watchdogs put their “stamp” on it. Clearly all these signs were quite tangible to all the regulators. One must notice that the figure shall NOT improve much in 2013 and 2014. Thus this year 2012 was really a bonanza one.

Now one wonders : “how much did the bank really make in all this?” The earnings line mattered less than the actual visible strengthening of the balance sheet as the former part tried to show. This account here is further evidence of the very continuous and close involvement of regulators all along the process. This balance between “tangible equity” and “other assets” not only match with the actual history of mergers and acquisitions that Jp morgan was involved with. It also matches with the recent history of financial markets, focusing on the credit derivatives, on the fundamental liquidity and capital issues that the banks faced since early 1999 at least. It obviously is also intimately connected to the financial crisis of 2008 and the seeming bonanza deals that Dimon got that same year. Truly this account sounds as counter-intuitive as the future “London whale” would be.

A total number of $52 billion has been suggested already although it was not a conclusive figure. One can get an overall proxy number of the gains that the bank obtained through the “London whale” scandal by looking at the evolution of the difference between 2011 and 2014. Keeping in mind that in the end of 2011, some $20 bln in off balance sheet position were at loss and not fully reported, some $4-5 billion liquidity reserve should have been taken just for the CIO “tranche book” that was to be taken down, one should set the “tangible equity – other assets” to “zero” rather than take for granted the $13 024 million paper gain of late 2011. Thus, adding the $13 billion of 2011 to the “visible” $25 billion of 2012, one can reckon that some $35-40 billion will be created in tangible capital in full synch with the London Whale mechanism in the course of 2012 alone. One can therefore safely assume that the scandal itself brought at least some $25 billion in tangible capital gains.

One should also consider then that at the end of 2011 the traded share price was way below the book value. Thus one even may go as far as saying that in late 2011 the stated $13 billion of excess value in “tangible equity” versus “other assets” was NOT deemed “reliable” at all in the markets then. Thus the “London whale” scandal really was a “catharsis” in many more worrisome aspects. It feels indeed that the “books and records” were plain wrong at year end 2011. Some rationale argument would corroborate this conclusion: a $4-5 billion liquidity reserve was missing at CIO for the “tranche book” and some “eliminated in consolidation” skew exposures were at a loss and unreported as such. Thus, including the recovery of the traded share price versus he book value, one can fairly conclude that through the “London whale” event Dimon generated $52 billion of tangible value for shareholders and regulators altogether. And what about the massive fines? Well, by the end of 2014, the bank had paid $1 billion for the “London Whale” and another $13 billion for other pending litigations. So, given that the quarterly earnings were barely affected, one should easily take the “extra” $13-14 billion of fines and add them to the gains of the firm. Thus this is likely a gross total $65 billion of tangible capital that Dimon generated through the “London whale” event “before expenses” on would say. Now if one remembers the many provisions that the bank had accumulated to the tune of $21 billion by 2014, one may conclude by saying that the bank even made $50-60 billion “net of all collateral costs”.

Now one may say “wait a minute! The markets rose, the bank balance sheet kept increasing, the asset prices rose altogether if only because the central banks sponsored it with their QE policies, the S&P 500 rose and thus the book value of jp morgan had to rise.” Thus these gains may have been simply due to the macro economic improvement or else the effects of QE policies. There are 3 arguments that will be developed to support that this overall $60-65 billion figure was not contingent to those factors reflecting the economic general improvement. First, the following table will show that the increase in the tangible equity was specifically tied to the timing of the “London whale” scandal, not the market rise as such. Second the bank had actually to deleverage its balance

57 sheet between 2012 and 2014, despite the apparent increase on paper. The short story here is that central banks used their QE programs to act as a pro-active clearing agent on tri-partite repos. This resulted for Jp morgan in a significant increase in “deposits with central banks” and a corresponding increase in ‘client loans’. Once adjusted for that change, the balance sheet slightly decreased in fact. Beyond that effect, the bank had to dispose of other activities at the time like the commodity market-making business. Third this analysis is based on the “book value”, not so much the “traded share price”. Yet if one looks at the outperformance of the traded price versus the S&P rise and adjusted for the historical “beta factor” of the Jp Morgan share price versus the S&P, one arrives at a similar order of magnitude that IS recognized by the market players actually. Thus here, if one accepts the idea that the book value anyway reflected the economic context for Jp Morgan, the outperformance of the traded price versus the book value reflects at least a $50 billion value added in the eyes of shareholders (all other things being held equal).

Let’s start first with the “timing” question. The table that follows will display the earnings history next to again the tangible equity versus the “other assets” and with now the story of the balance sheet size and the rise of this tangible equity. The ‘cash due from banks’ figure will be displayed for a specific purpose too:

One can observe that the total earnings line for 2012 at the top of the table looks like a toppish figure in 2012 that is only matched by the one for 2014, once the bank is rumored to be out of the woods for good. This looks like a spike of only say $1-2 bln versus the average yearly profit line that stands at $19-20 billion. But if one goes below on the “cash and due from banks” one notices that the bank is flushed since Q3 2011 with an extra $30 billion of cash that will only vanish in the course of 2014. This by coincidence matches with the fact that many off-balance sheet derivatives exposures will expire or be fully re-consolidated by the end of 2014. On the line below, one can see that indeed the total value of assets grows and thus the balance sheet grows apparently so. But as the last line shows, the “tangible equity” made a jump across 2012 that ended only in 2014. The chart below will plot this increase against the S&P 500 evolution. The year 2012 clearly stands out as the tipping point. The chart tried to keep similar scaling factors although this is just for the sake of clarity. Indeed, as per the argument, the tangible equity should move proportionally to the overall change in prices of assets and liability. Thus, the share of tangible equity should NOT change as a proportion of the total value of assets, unless something occurred. The chart below highlights the end of 2006 when the economy was peaking.

As one can see the surge occurred in 2012, not 2013 or 2014. Then as the table shows here, the ratio of tangible equity was at 4.1% for 2005, 2006 and 2007 with an S&P 500 at around 1400. At similar levels and no worries about the revaluation of “other receivables” and “other collateral”, in 2012, the ratio had jumped to 5.8%. Next, while the S&P 500 would rise by close to 20% in two years for 2013 and 2014, the ratio stays almost the same.

This is what the chart below plots. One will notice that the ratio started increasing in 2009. This shows that there was a long term strategy of Dimon behind that. But if one remember the SFAS 107 overvaluation on 2009, the doubts pervading the earnings of 2010 given the concerns of the regulators and the missing reserves for 2011, it is really only 2012 that the “London whale” scandal allowed Dimon to reach his long-term target so cathartically. Dimon’s job was “done” in 2012, not in 2013 or in 2014, even less so between 2004 and 2011….

Now the second argument comes that indeed “the S&P 500 grew, the economy grew, the balance sheet of Jp Morgan must have grown right?” Well it did on the face of it. But in reality, once the effect of the regulatory change relating to tri-partite repos is removed, the firm actually had to reduce its size. This was not significant at all despite the headline news made around the disposal of the commodity trading business. This was more a gesture than anything as the following table will show. Yet Jp Morgan did have to keep the same size while the rest of the world was growing. Size matters… This was a big hidden cost for the firm, mostly an opportunity cost versus its peers….Hard to quantify…But a cost for sure in a growing economy and bullish markets…

The effect of this change in tri-partite repo treatment caused a massive increase in “deposits with banks” and a corresponding increase in “fed funds purchased….”. Once the asset size if adjusted for what was not actually a genuine balance sheet growth but just a re-statement of the tri-partite repos overall, one gets to a different picture:

It is remarkable to look at the evolution of the total asset value versus the evolution of the total of the group (“deposits with banks” and “fed funds purchased”). The firm describes the “deposits with banks” as deposits with central banks mostly. “fed fund” refers to federally subsidized funding programs. Once one removes this group from the total assets, one realizes that Jp Morgan anyway barely grew or shrank actually since the end of 2008 up until 2014. Was it here really the free choice of Dimon here? One should doubt it if Dimon’s whole career is a guide….

In the meantime the economy crashed shortly and recovered. In the meantime too the S&P 500, rose from 800 points to more than 2000 points setting new records for 2 years on the follow after 2012….The tangible equity at Jp Morgan did not follow the same pace: it jumped with a 2-3 years delay. Thus this creation of “tangible equity” was NOT so much of a byproduct of the general economic environment anyway for Jp Morgan. This is what is displayed on the line titled “cumulated tangible equity created”. One may argue that “it take time to generate tangible capital”…. Once again, on this matter of creating tangible capital, one can see that such was the strategy of Dimon in 2011: to create some “tangible equity” as pictured in Dimon’s slides of September 2010 already (see the attached file on the website for the slides in question here). But as it was shown the attempt was successful ONLY through the “London whale” event actually. Before it had only been unconvincing at best… One can spot that the calculation here grants “only” $20 bln of capital creation for the sole year of 2012. But again this assumes that the bank record was clean at the end of 2011. If such had been the case, then one wonders why the share price traded at a 20% discount to book value…..

Thus overall, here one sees an adjusted $72 bln of tangible equity that was created with the year 2012 being THE turnaround year in the context this multi-year plan that had remained mostly unsuccessful likely from 2009. This process is well pictured through the following chart that plots the cumulated “tangible equity” being

60 created versus the actual adjusted size of the assets (removing the deposits at central banks and federal funding related assets). As one will see on the next page, from 1999 to 2009, the bank will grow its adjusted assets from $650 billion to close to $1 800 billion but there will be no “tangible equity” created as a share of these assets (see the “zero” line). The “take off” in tangible equity starts in the end of 2010 by a coincidence and speeds up in 2012 while the adjusted assets do not move.

Thus in this cumulated tangible capital creation of some $70 billion in total, nothing was certain by the end of 2011, $20 billion popped in 2012 for sure and most likely $25 billion more popped up at last as the plan would only succeed through the “London whale” scandal…..thus the chart below plots in a way that the “London whale” scandal did act as a “catharsis” for a tension that had lasted since 1999….This “catharsis” as Artajo and Drew discussed on April 17th 2012 was to cure the problems deliberately for good in 2010 as they were lingering since late 1999 at Jp Morgan if history is a guide….

Now the third argument should be that, “still” the economic environment was good. Granted, but then the bank share price should have traded then in line with the S&P 500 and its long term historical beta factor. Even better, the share price should have mimicked the evolution of the book value that clearly rose over the period. The table below shows that market players clearly perceived in a magnified way how beneficial the year 2012 was:

Markets saw a $30 billion gain for 2012 in the share price relative to the book value. The cumulated effect of the 2012-2013-2014 reaches more than $80 billion. And as of 2017, it does not turn out that the share price of Jp Morgan was overvalued at 110% of its book value at the end of 2014, right? Thus markets can be wrong but here they did not miss the bonanza gain apart from the moment of ambiguous panic that Dimon conveyed in his statements of the time.

Thus whatever angle one takes, either “growth”, “balance sheet”, or “markets” the gain is always of the $50-60 billion range at the least “net of collateral costs”. The year 2012 was the profitable “cathartic catalyst” and the “London whale” was the event that brought up $25 billion or more as it came right when the bank would at last achieve a decade long awaited clean-up... Thus the improvement was quite tangible for anyone to see, including all the regulators scrutinizing this CIO and this synthetic “tranche book” falling in shame. The coming table below summarizes this part in 4 stages ( in red the troublesome figures)

One may have trouble reading this table above, so there is a link that is available to the original excel spreadsheet on this website “Summary of the gain of JPM. XLS”

As one can observe above, the troubles of the bank started in late 2008 with a couple of worrisome evolutions that were structural: ever higher “intangible” capital, lagging “tangible equity” versus “other assets”, lagging share price versus “book value”. The issue was there already in 2002…One can next notice the efforts in the period 2009 to 2011 to improve the picture. The “tangible equity” appears to be created in 2010 and 2011 but in a way that is neither convincing the regulators nor the markets in general, witness the visible drag on the share price versus the book value or the regulators official concerns of the time. Then in 2012, the “London whale”

63 event will surge, fueled by “orders” repeated to Iksil that were quite strange still and were as many flashing red lights. Yet the orders will turn all firelights to “green” after the “tempest in a teapot”….

There was here a long term plan behind the “tempest in a teapot”, addressing a crucial long term issue in fact. This is NOT by chance and NOT in an unrelated fashion that the “London whale” scandal emerged then in early 2012. The orders went against basic intuition but matched a well thought off strategy for the whole firm. No wonder now that Iksil’s alerts were very well flagged and understood, but they would not move the needle for Dimon who was ALREADY addressing the issue with regulators on THE very same issues that Iksil kept alerting on. Now one may argue that the bank certainly did NOT want the scandal to arise like this. Did the bank then want the loss to balloon at CIO knowing that regulators would complain as CIO was meant to protect the bank and its depositors in principle? This makes no sense, does it? Did the bank want to shrink the skew to zero on IG9 10 year precisely as it was THE benchmark for pricing the basis risk in CDS markets? Well the bank could afford to wait a little more time for the remaining skew positions to expire. But could the regulators wait any longer as their mandate empowered them more and more to take action? Did the bank anticipate then that rumors of manipulation would arise and therefore had prepared a “trader” to fall for that well ahead of time- just in case? Well that “london whale” trading scandal was quite a convenient “incident” for regulators who had pushed for a strong “Volcker Rule”. Did the bank do all that also to maximize its profits every quarter on top of the huge projected gain it saw coming anyway through this internal collapse? Well the bank obviously made a unique and predictable fortune if history is a guide. In short, did the bank manufacture this “trading scandal” on purpose? If it did it was not unknown to the watchdogs watching. Did the bank have the choice actually when considering the missing liquidity reserves for CIO for manufacturing this “trader tale”? It did have a choice at the end of 2010. The VaR story surrounding the “London Whale” scandal provides further clues to answer all those questions above.

One however has to remember that a lot of issues remain unsolved around the VaR management of Jp Morgan through the event itself. Those issues were flagged by the US Senate report. They did NOT result from CIO traders actions or even CIO managers actions. They were admitted lies and misrepresentations by the firm. They have NOT been addressed to date by any regulator, ie none of the executives involved have been held accountable for that on a personal stand. A short story of their role will come as a way of conclusion for this memo. But first, it matters to see better how and why the loss grew so much, so fast in this “tranche book” of CIO in the first quarter of 2012 and on the follow. This will clarify how the bank did sail through the scandal within the walls of the fortress. It did maximize its profits in a surgical manner.

But before looking into the way the losses snowballed into the CIO book itself, before and through the “London Whale” articles, it matters to secure one central point.

And the point is: “the loss at CIO was an integral part of the plan long designed and commanded by Jamie Dimon since 2010. This is this plan here that generated these providential $50-60 billion of tangible capital gains. In the way this gain showed it could only be a basis risk that had gone away”.

So far it has been established that there was a massive gain happening simultaneously, that it had been spread indeed across several segments of the firm but all through the use of “hedging” techniques that were NOT “specific hedges”. More this tangible gain touched upon “tangible equity” versus “other assets” which, for the latter, is a class of assets containing leveraged exposures that are NOT necessarily captured through the mark to market process. The reason is simple: the prices are uncertain for an underlying market that is almost non- existent. More this asset class is structurally NOT a classical receivable or payable amount. The gain has been given no explanation by the bank whatsoever although it visibly materialized in the second quarter of 2012 and got bigger and bigger until late 2014. On the surface, it may look rather unrelated to the “London whale” loss of

CIO on credit derivatives that were in Mark to Market and allegedly served a “hedging purpose” although undocumented.

Many hints were made already about the spontaneous connections that did exist between the “tranche book” of CIO and these net $60 billion tangible gains. But was this “tranche book” at CIO a hedge really going against these massive gains? If true the loss at CIO made total sense on this “tranche book”. If true, the bigger the loss on the “tranche book” the bigger the net gain for the bank predictably so. Where is the link? The bank has always been more than ambiguous on the matter. People by the way always doubted that it was a hedge precisely BECAUSE the bank NEVER disclosed what kind of assets/liabilities this “tranche book” of CIO was actually protecting on the balance sheet. As shown it would not have been such a great deal of mystery due to the very nature of “other assets”. It was not so hard or so incriminating for the bank to actually flag the connections mentioned here and further explain how the connection effectively worked inside the firm since 2006. Many huge other amounts would corroborate the story. Many evidence exist to document the story as well. The stakes understandably were as huge as one can imagine, and not only for the bank or its shareholders here. The $3 trillion in notional exposure of “identical underlying” instruments that all point to the “basis risk” are spread between “on balance sheet” and “off balance sheet” exposures. This “gap” is what the “tranche book” at CIO was primarily meant to protect the bank from. But today the doubt prevails since the bank would never clarify the situation. This is another puzzle as here the bank really hurt its reputation in a self-inflicting manner.

Yet one should concede already that at least the CIO “tranche book” was directly related to the “basis risk” of the bank by its original mandate as stated in the NBIA of 2006 by the bank itself. One should also admit that “other assets” were directly exposed to this “basis risk” by their very nature as the bank describes them. This is the source of the link, not the link itself. But one then could rightly argue that, one, nothing proves that the tangible gain was done on this “basis risk”. And, two, one could also argue that the loss at CIO was not a fatality carved on the roadmap of the plan as it had been designed by Dimon. Two evidence link directly Dimon to those 2 arguments, proving them wrong. The first argument was that the “tangible gain” was not necessarily realized on “basis risk”. The evidence that proves it wrong is the September 2010 presentation done by Dimon himself. Some key slides will be displayed below that show how Dimon planned to put in “run off” credit derivative portfolios in order to free some capital as the new Basel III standards approached. Here the plan was to effectively collapse “tranche positions” across the firm and thus collapse massively expensive “basis risks” when they were expressed in capital terms. That was clearly expected therefore to be quite a tangible capital gain. And that’s what happened actually. But let’s first also prove the second argument (ie “the CIO loss maybe was an accident”) wrong before displaying the slides themselves. The proof against the second argument is found in the string of emails dated April 5th between Drew and Dimon himself. Here Drew explains her decisions, the loss and the fact that the earnings will NOT be impacted despite the $500 million “DRAWDOWN” that was NOT addressed in full. And Dimon here will want to know whether Drew’s coming communication is “related or not” to “our credit exotic wind down”. Dimon himself saw a direct relation anyway between the “drawdown”, ie a projected loss on the “tranche book”, AND his “wind down plan”. The fate of the “tranche book” of CIO was naturally related to this long expected tangible gain. BUT Dimon wanted here to know actually whether Drew in person wanted to talk about the connection as she spoke of the current growing loss and the “DRAWDOWN”. To this Drew will reply that Dimon probably thought of “his” internal collapse of tranche positions as planned with the IB involving the loss to snowball at CIO. And she will state that she will NOT talk about that. No, she will talk about the fact that CIO highly confidential positions had been leaked to the markets, had lost big as a result, and that hedge funds had been tipped and that Hogan was “helping” CIO. This alone shows that the link “CIO loss-Plan” was directly managed by Dimon, not Drew. It shows that the loss at CIO was expected as such but was NOT to impact earnings (good!- who could check on that but the regulators?). It shows that Drew was just following Dimon’s instructions although this evolution closely involved the IB and hedge funds then. As to the “basis risk” connection, one should usefully read again

65 the conference call between Artajo, Macris and Pinto of March 23rd April 2012. One should also read the call between Stephan and Artajo on the same day where this is Stephan here who describes so well the “basis” risk exposure that CIO and the IB managed together. One should also read the call between Artajo and Irv Goldman of March 30th 2012 where Goldman calls on behalf of Drew who seems to want to know about the “bid offer attribution” details AND mostly the imminent projects of holidays of Iksil.

It is worth thus repeating the key fact : “the loss at CIO was an integral part of the plan long designed and commanded by Jamie Dimon since 2010. This is this plan here that generated these providential $50-60 billion of tangible capital gains. In the way this gain showed it could only be a basis risk that had gone away”.

To secure this with available evidence now let’s review first some key slides of Dimon of September 2010, 18 months before the surge of the “London whale” tale:……on Slide page 17, Dimon lists the “things they will watch”…..they are circled and it is really hard to imagine that the “tranche book” at CIO is not targeted for a “run off” here along with “credit hybrids”…..

The next slide on page 18 is worth a read too in the attachment. It is not displayed here as it will NOT be easily readable. But this slide on page 18 shows that many regulators, especially the Federal Reserve, have quite a close look at all this at the time. One really wonders how Dimon could just lift his little finger without an army of watchdogs watching that out….

The slide on page 22 depicts the “impact” of the recent regulatory reforms so far…. The bank is reviewing all its potential “proprietary trading” activities…..Regulators do the job with the bank here since the Volcker Rule is ALREADY known…

The issue is being addressed already about “proprietary trading” and one really wonders what any regulator could discover about the “tranche book” at CIO or Iksil here in 2012, especially after the February 13th 2012 letter of Barry Zubrow. The case was closed already by then, 2 months before the tale would take shape in the press.

More, beyond the context of the Volcker Rule, there were other reasons to “take down” or “dismantle” the tranche book of CIO and they are listed on page 23:

As one could expect the bank reviewed the portfolios, the activities AND it also reviewed the products one by one. The considerations were driven by profits under the new regulatory constraint: stepping through a clearing agent that would de facto decide on the mark to market “mid price” for the bank. The bank here points to the ICE project. The change depended upon the liquidity of the product considered. For the il-liquid” products the prices would be affected simply because they could NOT be cleared by ICE as there could NOT be a consensus that would be reliable enough. It was the case for the “single name CDS”, the very components of indices like the IG9. Therefore, since the IG9 was still “deemed” liquid, the “IG9 skew” showing the difference between the IG9 index and its very single name CDS components, had to increase. These statements of Dimon here show that the creation of ICE made the indices cleared by ICE even more liquid than their own components which were NOT clearable through ICE. This is here a recognition of the very nature of the “basis risk” and the “skew risk” in particular. Dimon in his own words displays his personal knowledge that the ICE project will further increase the difference of liquidity between the indices and its constituents due to the existing contractual differences. Here the bank betrays its knowledge that in 2012 the IG9 10yr skew was expected to increase starting in 2011 rather than squeeze down to zero as it occurred so extraordinarily through the “London whale” scandal. The regulatory pressure put on Jpmorgan and other big banks to participate in ICE shows as well that an imminent widening of the iG9 skew was expected by the watchdogs starting in late 2010. This therefore also shows their puzzling silence on the IG9 skew going down to zero in June 2012 as one that is complicit with those market players who manipulated the IG9 skew for it to land at zero then.

More the second part shows that “High Yield and certain investment grade CDS” might have to be “moved” not “unwound”. Here is the “wind down” project by which the “tranche book” would be “taken down” through an

“off-shoring”. Dimon listed here the credit derivative products that constituted 90% to 95% of the “tranche book” of CIO. Yes this CIO book was naturally targeted to be “moved” as winding it down in the markets was precluded. And so was the “tranche business” of “credit Hybrids” at the IB right at the same time….

On slide page 28, Dimon here summarizes the strategy of the bank: “

Just fine! On page 30, Dimon put next a couple of sentences that must have sparked strong concerns among all the regulators. First, one can see that the minimum capital requirement under the future Basel III standards would be 7%. The footnote specifies that the rules are NOT finalized yet. It matters since this shows that the bank works hand-in-hand with all the regulators on the Basel III issues. Next the bank states that “currently” the capital under Basel III standards is at 9.6% (versus a floor at 7%, which is not quite a big cushion if the basel III model proves wrong in real life through the next crisis). But worse, the bank projects only a 9% capital for a key date “after taking known actions”, ie the end of 2011. Dimon here has plans for the end of 2011. Regulators are aware here, right?

This is what he refers to stating :

Yes the bank shall “take actions” that shall have to conclude by the end of year 2011. And this will induce a lower capital ratio for the bank looking forward under Basel III projective standards. The “tranche book” is targeted at the IB and at CIO. No regulator will even try to meet Iksil. If the internal collapse of synthetic tranche inside Jp Morgan is any guide, it was one target of those “actions” indeed but it would NOT be done on time as the “London whale” scandal itself testifies… By a fantastic coincidence the “London Whale” scandal would surge just few weeks later than “year end 2011” as a rumor in the markets would spread where this “tranche book” of CIO was targeted. But even then, had the collapse gone as planned, the capital would “only” have been at 9%, shrinking further the capital cushion. Thus those new regulations unveiled a capital plan for Jp Morgan that revealed that the bank was NOT such a “fortress” any longer.

But Dimon made matters even worse on the following slide versus his own statement of page 28 above…. He added here that he would not need or intend to issue stock (see the extract below). In fact Dimon will let the media know that he would instead start a massive buyback of shares. Was it a “mistake”? No. This is a plan that the Federal Reserve will approve and support surprisingly so.

And even worse, Dimon will state here that: “

One wonders here: “what was the regulators’ belief on this matter here?” Were the Basel III standards so safe while they were untested and actually un-finalized? Is Dimon serious here? It sounds like a genuine provocation, like not a “tempest in ateapot”. Well, the initial reaction or regulators betrayed a heightened concern. The Federal Reserve will prepare to launch no less than 2 thorough investigations on CIO for 2011. The FCA will meet a couple of times and issue its ‘close and supervision” letter on November 2010, targeting CIO and the ‘correlation book” in particular. The mood among regulators was suspicious as the MRA letter of the OCC in December 2010 proves it. They wanted to get to the bottom of it seemingly so, didn’t they?

Drew will reply with a “stern” answer in January 2011 indicating that everyone knew well what was going on… All right!…So much for the transparency as just none of those regulators will disclose much of their concerns of the time in the future. They actually will redact their original concerns as much as they can in their future public investigation reports. Still the US Senate Report spontaneously conveyed the OCC’s view that Drew then sent everyone back to Dimon and the Board. Was she really pointing only at her higher ups here with this “stern” reaction? One may assume that if “yes” she was actually pointing upwards in the firm, she would have sounded either overly “defensive” or “apologetic”, but not “stern”. She was NOT the boss of Dimon or the boss of the board members, right? Drew meant what she said in her stern reaction and regulators apparently concurred. They at least knew what she meant by that “stern reaction”. Was she saying somehow that regulators acted in a hypocrite manner? She will commit to better document the $360 billion investment process. BUT she will NOT commit to better document the “tranche book” valuation process or decision process. And regulators will let her do as she wanted…The regulators knew better than her anyway it seems, didn’t they?

The bank will send quite misleading reports on the follow of this “incident” about stress test limits breaches in January and February 2011 alleging a change on the “tranche book” that never occurred then. And paradoxically so, the FCA shall meet CIO on this “tranche book” but without willing once to meet with Iksil. The OCC will put on hold the follow up of its own MRA and the Federal Reserve will surrender its agenda to the LISCC committee. The latter obscure committee will decide that there are “other priorities” for all US regulators regarding Jp Morgan. Drew maybe was therefore pointing the finger back at the hypocrisy of the regulators. As a result, Jp Morgan will be praised publicly by the Federal reserve in March 2011 and be allowed to buy the shares back in mass while its CIO had lastingly breached its own stress limits....If that is the case that regulators then were just playing games then the whole public blame around the “London Whale” was really unfair indeed.

It is thus worth showing the slide in full to “see” the event that sparked all these gestures:

The next slide on page 31 is very interesting too as it connects straight all this to the “tranche book” of CIO…..

One can see above an expected $15 billion “capital generation” on “known actions” that were to be finalized in Q4 2011. But then in September 2010 the IG9 10 yr skew was NOT at zero. It rather was quoted at around 10- 12 Bps. It seemed just fine for Dimon then. What would change for the bank and Dimon in 2012 or 2011 that would force the CEO to wait until the IG9 10yr skew would be pushed down to zero so manipulatively? Rather than $15 billion, the ultimate gain would be of $75 billion, ie $60 billion more. Net of all the “collateral costs and fines” the bank will be better off in capital to the tune of $50-60 billion thus “net of charges”.

The other circles describe those “known actions”, ie “market risk-reduce IB & CIO positions” AND “RWA on 50/50 deduction; reduce IB/CIO securitizations exposure”…..This was clearly related to the “Tranche book” of CIO indeed. No other book at CIO could be the target here. The plan would be delayed and actually “terminated” in July 2012 with the “help” of quite well timed a scandal actually. This focus on liquidity reserves and capital provisions targeting both CIO and the IB at the same time is further confirmed on the page 34:

See “RFS” here? It is a reference to bear in mind for the future comments of the August 2012 statements.

There is here further confirmation that the “estimated notional impact” was expected to materialize “by the end of 2011”. This echoes the sentence of the UK CEO of Jp Morgan on March 23rd 2012, Daniel Pinto, who told Macris and Artajo that the IB was hedging CIO on the high yield market in 2011 before American Airlines went bankrupt. Neither Macris Nor Artajo would betray any surprise here. That was “strategic”, as opposed to either “opportunistic” or “tactical”. They could have been surprised by Pinto’s statement since by mandate it was the “tranche book” of CIO that was to “hedge” the IB, not the opposite. Yet Pinto described things that were done like this because that was as per a plan long designed by Dimon here. That plan was quite strategic with huge stakes at play with the regulators, with the shareholders and with the markets. The connection is now straight between “tranche book”-“CIO”-“basis risk”-“end of 2011”- “tangible massive capital gain” and the actual trades that would be done BOTH at the CIO and at the IB. It is made by Dimon himself and in front of all the regulators watching that closely for obvious historical reasons related to the last financial crisis of 2008. And Dimon further provoked their scrutiny with his statements….

Now it matters to prove against the second argument (“the CIO loss was not planned”) that Dimon saw the CIO loss as a well anticipated event that was under complete control as far as the earnings were concerned. The proof as it was said is found in the April 5th email chain. It is going to be reproduced here for a complete read.

“Senate report exhibits published in November 2013: Message 1 ----From: Drew, Ina---Sent: Thursday, April 05, 2012 05:58 PM---To: Dimon, Jamie; Zubrow, Barry L; Staley, Jes~ Cutler, Stephen M; Maclin. Todd; Braunstein, Douglas; Erdoes, Mary· E; Smith, Gordon; Peloo, Douglas B.; Bisignano, Frank 1; Hogan, John J; Cavanagh, Mike---Subject: CIO----I want to update the operating committee on what is going on with the credit derivatives book in CIO especially given a wsj article which will come out tomorrow. One of the activities in cio is a credit derivatives book which was built under Achilles in London at the time of the merger. The book has been extremely profitable for the company (circa 2.5 billion) over the last several years. Going into the crisis, we used the instrumentation to hedge mortgage risk and credit widening. Recently, in December, the book outperformed as it was positioned in for "jump" risk or default risk throughout the summer as a relatively inexpensive hedge for fallout from weak markets during the european crisis. The fourth quarter 400 million gain was the result of the unexpected american airlines default Post December 2011 the macro scenario was upgraded and our investment activities tuned pro risk, the book was moved into a long position The specific derivative index that was utilized has not performed for a number of reasons. In addition the position was not sized or managed very' well Hedge funds that have the other side are actively and aggressively battling and are using the situation as a forum to attack us on the basis of violating the Volcker rule Having said that, we made mistakes here which I run in the process of working through. The drawdown thus far has been 500 mil dollars but nets to 350 mil since there are other non derivative positions in the same credit book. The earnings of the company were not affected in the first quarter since we realized gains out of the 8.5 billion of value built up in the securities book. John Hogan and his team have been very helpful. I wanted my partners to be aware of the Situation and I will answer any specific questions at oc monday. Have a good holiday, “

Message 2

---From: Dimon, Jamie---Sent: Thursday, April 05, 2012 06:00 PM To: Drew, Ina---Subject: Re: CIO---Ok. Send me some info. Also how does it relate or not to our wind down credit exotics book?

Message 3

“From: Drew, Ina [email protected]: Thu, 05 Ap' 2012 22:08:57 GMT---To: Dimon, Jamie [email protected]: Re: CIO---If you are referring to the wind down in the ib credit exotics book, it is separate. Achilles and I targeted the CIO tranche and derivative activity as a reduction item (I specified in last bus review) due to the high rwa it draws under basle III. We have also had issues with QR that have made the rwa outcome less predictable. However we are working with Ashley and Venkat to see IF both the IB and CIO positions could be moved out into the winters fund. I have been assessing the trade off between P&L and RWA for the second quarter. I can go over all the technicals with you at any time. I wanted to this week but understood you were on vacation.”

One point about “holidays” is to make here. On April 5th 2012, the very eve of the seminal article on the “London Whale” legend, Drew is apparently puzzled to learn at the last minute that her direct boss Dimon is on holiday “apparently”. She was not told she says. She is not quite sure actually this is true given that this is all about the plan of Dimon as exposed in the slide of September 2010. This very late awareness of Drew about Dimon’s holiday right when the first quarter of 2012 was ending is quite surprising as such. First the CEO is unlikely to be off when a quarter ends. Second Drew displayed an obsessive attention to “holidays” at the time… Indeed the call between Artajo and Goldman on March 30th 2012 ( in the US Senate report exhibits) proves that Drew actually had quite an obsessive focus on holidays and specifically on Iksil’s plans on the matter right then. So it appears that Drew had been checking compulsively “who” may be off in her teams few days ahead of the articles but she had not checked with her own direct boss while she had so many topics to raise with him….This cannot be here a coincidence as Drew was definitely pro-active. Was she just truthful in her email with Dimon repeating her apparent surprise about Dimon’s time-off? Or maybe that this was just a sort of coded message between Drew and Dimon about the surrounding hypocrisy that prevailed since late 2010. Yet, even if he was on holiday apparently so, Dimon was NOT moved at all by this loss and the rest that Drew pictured in her email.

Thus it matters to repeat a third time that: “the loss at CIO was an integral part of the plan long designed and commanded by Jamie Dimon since 2010. This is this plan here that generated these providential $50- 60 billion of tangible capital gains. In the way this gain showed it could only be a basis risk that had gone away”

The question that is going to be addressed on the follow is therefore:” why did the bank silence completely the so profitable outcome of this plan of Dimon since the beginning of 2011 and onwards until today in 2017?”

The items below would further document the fact that this quite tangible massive capital gain was indeed tightly related to the snowballing losses of the tranche book of CIO in 2012….The top executives conveyed instructions inside the firm and made public statements that would make the loss to be that big. Otherwise, as some elements suggest especially after the seminal articles, the “tranche book” at CIO could not have lost these many $ billion. Was the senior management of Jp Morgan making additional mistakes in finalizing the “wind down” plan” and communicating to regulators or the public? Not so much actually….One should indeed remember the repeated concealments of the top management about the 10-Q VaR starting in January 2012. That was intentional all along and was necessary to make the scandal be possible ahead of time. It also on the side

74 corroborates the view that the restatement based on initial price differences should never have been done by the bank or been “approved” by regulators. Indeed the firm specified in its VaR section of the 10-Q of Q1 2009 (pages 71 to 76) that some “IB correlation risks” were NOT captured in VaR. This only corroborates why, although fully in mark to market itself, this “tranche book” of CIO collapsed in a much broader operation where other parts were NOT accounted for as per a “mark to market” process. On the same VaR section the firm also specifies that the VaR analysis is based upon “market related revenue” across business units. This word R-E-V- E-N-U-E in the VaR section indicates that the bank based its VaR aggregation of CIO, RFS and the IB in particular from P-R-I-C-E-S standing for “market consensus mid prices”. Once again, consistently so with the collateral netting process applied to “legally enforceable” contracts, the firm here stated that it did reconcile and adjust any price difference between the business units as far as the “tranche book” of CIO was concerned. The other items below would show that the whole infrastructure of the bank was based upon this basic common sense feature…..

SFAS107 history and “other collateral” amounts finally settled or “discontinued”…. PWC vs JPM: the ongoing mismatch finally solved in 2012 and onwards….. Projected Cost of assets and liabilities finally lower after the scandal….

After showing the massive gains and its direct connection to the CIO losses on the “tranche book”, it matters to show whether this loss was scrutinized, was fully controlled and was actually a wanted windfall ultimate gain for the firm.

3- How did the loss spread into the book in 2012?

There remains at this stage a significant possibility that maybe the bank had just been “lucky” or “opportunistically smart” here sailing through this scandal to achieve its long targeted huge gains. The fact that the “tranche book” was in the cards for this “wind down” plan turned out to be a problem. Yet it should NOT have been the case, a problem. So would be the official tale of the bank in 2017. What was the problem? ‘Mistakes were done, ok?’ so said Dimon in August 2017 to summarize the official story during an interview with CNBC. No regulator would comment here. Conspicuous silence again… Endorsing that scenario would mean that maybe the bank simply achieved its target “despite” the CIO mistakes, optimizing gains every quarter on the follow still … thus being successful “despite” the scandal, despite the admitted “management’s mistake as a group”…. Achieving in fact a $50-60 billion historical unique tangible gain in capital straight net of costs…..

One should just assume then that the top executives were sort of “long term smart-greedy and acted with perfect integrity at the end of the day”. This assumption would plainly justify why the bank was blamed, settled in October 2013 claiming “unfairness” and why on the follow not a single executive would be charged on a personal basis by any authority investigating the case still. The latter knew even better therefore… But they claimed all along a crass “unawareness” didn’t they? This is quite a striking outcome as no less than the FCA, the UK Serious Fraud office, the US DOJ, the US SEC, the US CFTC, the US OCC , the US FBI, and the US Federal Reserve were on the case here “in hindsight”. If such is the final outcome, this will carve in stone an historical jurisprudence as the evidence abounds be that for top executives or be that for any employee down the chain command.

Were the senior managers of JpMorgan so candid? The following part will show unfortunately that there was no room for random coincidence or “lucky/Unlucky opportunity” between the ballooning losses in this book at CIO and a string of key events that occurred within the walls of Jp Morgan. The following part thus has material

75 consequences with regards to the jurisprudence that the “London whale” case will establish for the future about senior executives as well as for any employee that becomes “targeted” by the authorities whatever their reason was, is, or will be….

The former part had shown how, given the high stakes lying behind the “wind down” plan of Dimon, the alerts from Iksil could be well understood, they would weigh almost nothing on the balance. Fine. Who would care today? Employees served as scapegoats. Their lives were destroyed. As such this does not mean that the top executives did anything wrong other than “their mistake” which was the one leading to the scandal actually. “Bruno…-my personal view- He is not the one to blame” Dimon would add finally during the same CNBC interview of August 2017… Who would have guessed that since 2012? Was Dimon the CEO and board chairman of JpMorgan then? He was not listened to it seems. But by whom was he not listened to?

This part here will show that the many alerts of Iksil to stop trading were ignored with a specific purpose which annihilates a lot of the candor that could be granted to the senior managers: the losses at CIO in this book were to be maximized following quite a tight agenda of theirs while maintaining a semblance of liquidity for those “tranche book” positions. This is about “Smokes and mirrors” here, not outstanding “adaptive” skills…That set of instructions from Drew since December 2011, and “covered by Jamie” all along, remain in the shadow still today in 2017. The orders in question would be issued and actively supported temporarily so then by all the senior managers of JpMorgan. All this was done in order to make believe that the “tranche book” positions were still “deemed actively traded”. They had a purpose here back in December 2011.

Iksil advised in writing instead at the very same time in January and early February 2012 to “take the full pain” announcing a “liquidity trap” and an easy “drawdown of $300 million”. Iksil also advised to let “this book die” as it was caught in a “liquidity trap”. He did try to “do something about it”. And big things happened on the follow. Drew was “not worried” but that was a gesture of hers: Artajo would be demoted right then and Macris was quite worried. That series of orders of Drew was here, and the demotion of Artajo as well, for any outsider like regulators’ staff to see. How could they fool themselves so much then pretending that they “believed the book was being taken down” quietly, breaching their own routine duties right then by NOT opening any of the reports that they demanded normally? Iksil could yell in the desert he would have been better heard.

The reason for that last short term delay in the “wind down” plan and this unbelievable “blindness” among the watchdogs in Q1 2012 thus was to allow this temporary period of active but artificial trading activity. Then the “London Whale Marionette” would be positioned to take the fall for all on the front-stage under the spotlights of the show being played in the markets: “JpMorgan IB versus JpMorgan CIO”. This was well prepared since December 2011. The articles got to press in April 2012, ie 4 months later. The goal was straightforward: the “tranche book” at CIO hedged the bank against an increase in the “skew” and the plan easily projected a maximum gain if the “skew” went as low as zero. That mechanically implied a maximum loss on the “tranche book” itself. Any point in time since 2010, the IB and CIO could have achieved this wind down operation as quickly and seamlessly as they did in a matter of 2 weeks in June 2012. They only needed the “go ahead” of Dimon to execute a seamless preliminary transfer of the “tranche book” of CIO to the IB. That is what they did at the top of the bank but only in late June 2012. Since 2010 the bank anyway had enough capital to make the transfer from CIO to the IB and next proceed at the expense maybe of a new liquidity reserve. This reserve would have been massive but it would have been needed only for a well known period of time. It may have forced CIO to divest some assets but it would have been not so expensive in the end given the projected $75 billion gain before collateral costs. That part was easy to justify or explain to the markets or to the watchdogs.

But instead of being straight in late 2010 the bank would wait for the optimal IG9 10yr skew level and then just did the easy preliminary transfer at this specific spot in June 2012. The media pressure or the regulators pressure

76 or the political pressure or the FBI pressure clearly had been resisted in full through that period between say October 2010 and June 2012. Thus the only catalyst left that launched the very final stage of the wind-down plan is this IG9 10yr skew “price” itself. This is just another inference.

One wonders still what the “practical thing” was that would justify the delay about the last stage of the “wind down” plan from December 2010 and onwards. It must be a “thing “ that was closely related to the need to anchor the IG9 10yr skew at zero and to push the corresponding predictable loss in the “tranche book” of CIO at its apex. Such was what the wind down plan conveyed: “The transfer of the “tranche book” would NOT be transferred “off-shore” until the target in skew was achieved”. To be clear, the “tranche book” would not be “wound down” before the IG9 10yr skew went to zero. But that very, very likely induced moving the markets where they should just NEVER trade on IG9 10yr skew. People would easily suspect a manipulation here to make maximum profits. How could the bank expect to escape the suspicion? Better was to plan a good old trading scandal at one stage to settle the markets at an IG9 10yr skew at “zero” for a while with a nice “hook” and a big “whale hanging on the hook” for everyone to see.

The ideal way here was to make believe that “CIO had had flawed positions and a stupid strategy”. Of course that could only be the fault of a “trader”: a mistake for another mistake, so was the tale…. That was just one more decoy in the fishing party. That would not last for very long, but long enough to finalize the ”wind down- off-shoring- collapse” plan with regulators. The latter surely were not blind after June 2012. This is how the $60 billion gain described in the former part could be secured in practice. But in early 2012 time really was running out: resisting all the other pressures was tough. And yet the bank did thanks to the exceptional “admitted negligence” of just all the watchdogs in a quite exceptional synchronized move on their part towards “perfect inefficiency”.

That gross setup indeed would induce further “mistakes” among the senior managers. The issue around the skew was central for everybody involved in this “wind down” plan of Dimon, inside and outside the bank. The orders came from the top. They were quite specific and self-conflicting. But the “mistakes” were not located around the trading issues actually like volumes, execution, notional amounts, estimated loss or price differences (against the official tales).

It was easy to keep the “tranche book” loss itself under full control. Iksil had long documented the situation since 2007 actually towards people who had directed him all along to do so. That was mostly his job at CIO since he had joined in 2006. There was no mystery in that dedication around the skew as, by design, this “strategic hedge against a liquidity crisis” based upon credit derivatives was targeting the “skew risk” in just all its strategies first and foremost….Thus in early 2012, the case was clear: as the “tranche book” of CIO loses money, the bank makes even more with certainty (remember the $3 trillion of “identical underlying”). The “hedging” activity of the IB to “hedge the hedge at CIO” was really fine tuning anyway. Every basis point of IG9 10yr skew may bring a gain for the bank of $1.5 billion net of the loss that the “tranche book” would record anyway (this figure can be inferred from the risk figures provided in 10-Q reports next to the identical underlying table shown at the very start of this document- see also the first simplistic example at the start of this document)…

This part will show how granular the knowledge was of the “tranche book” risks with regards to any aspects related to trading as mentioned above. The tables and charts that will be displayed here are just “proxy reconstructions” of the works that Iksil performed and communicated at the time repeatedly in CIO. As such they convey the basic conclusions that anyone at CIO could see and understand. The manipulations of prices in the markets were clearer and clearer every day since January 4th 2012. The specific targeting of just every single

77 position of the “tranche book” of CIO was also clearer and clearer by the day all along the first quarter of 2012. The phenomena were quantified in detail and elevated “all the way up”.

This part will convey the essence of these phenomena and the associated alerts of Iksil that started in 2012 no later than the 11th January 2012. Then David Goldenberg from CSFB (worked for Andy Hubbard, the husband of Stephanie Rhule at Bloomberg) conveyed the first clear indication of a targeting AND a manipulation of IG9 prices due to skew based interests. The coming charts and comments are just recollections that were started between May 2012 and early 2013 while the memory of Iksil about his day to day actual job was still fresh. It was easy to remember actually. Iksil’s job in 2012 had mostly consisted in analyzing precisely what was going on around this IG9 10yr skew. But the analysis would always be a bit incomplete. Indeed Iksil could not access the valuation engine tool at CIO. The engine required a password that Iksil could have had but never used in practice. Whenever Iksil had had a password, the IT systems of the bank closed the access of Iksil after 30 days of “no use” because Iksil simply almost never used the password for quite long periods. Iksil had given up on trying to access valuation engine tool of CIO since 2011 at the latest. Therefore Iksil had NO direct access to the prices or to the estimate P&L results in 2012 apart from the emails that Grout was to send across the firm every day.

That was way enough actually. Iksil was NOT among the ones who ran the estimate P&L process against the repeated pretence of Jp Morgan and his close colleagues. He was to eventually help in case of emergency. Otherwise Iksil would be simply asked for his intuition and market practice by Grout or Artajo at specific times in the day for Grout and Artajo to process the estimate P&L report. That was the valuation setup wanted by Drew, Macris and the one setup that had been implemented by Artajo since 2007, way before Grout would join CIO in 2010 to replace his departing predecessor. They had quite sensible reasons to operate like this.

Iksil therefore relied all along on proxy calculations of his and relied on his own intuition as to which prices to select for his own duties. That independence was welcome so that Iksil, Grout and Artajo arrived at their own views first without being too influenced by the others. That independence also was a key part of the process designed by Artajo back in 2007. It was surely time consuming and as a result the figures of the time remained well carved in Iksil’s mind. Thus the prices and drifts in relative pricing terms may be not all 100% accurate but they display the right context for the losses anyway as Grout or Artajo supported Iksil’s findings all the way. The figures below may thus not be exactly like the ones that were communicated back then by Grout or Artajo inside the firm. But the big picture is the one anyway that Iksil elevated in due time in the CIO and that Drew elevated loudly as of March 23rd 2012 “all the way up”.

The main scheme of the text that follows is:

a. the DTCC/ICE open position data shows that the IB was in front of CIO all along and adjusted its own positions to CIO changes in the course of 2012. It had to do that on behalf of JpMorgan and in direct relation to regulators concerns about the market share of the bank in CDS indices. b. the loss analysis shows that the bank maximized the loss at CIO on skew related positions while trying to reduce the total loss for CIO alone. Sounds very much like a well planned loss for the “tranche book” while the earnings would not be impacted (see Drew to Dimon on April 5th 2012, right before the articles went to press)…. c. The positions would have recovered from 2013 onwards in full had CIO kept the book despite the alleged “flaws”. The recovery of the losses was quite predictable and would have happened in practice anyway. This also was known very well inside the bank in early 2012. The collapse thus occurred as soon as the projected maximum loss point was reached for skew related exposures at CIO in June 2012.

Looking at the 3 points above, one may wonder what had been left to chance here: the positions of CIO were scrutinized and hedged “live” by the IB if needed. They were valued for 90% of them by ICE daily. They were to lose but in a controlled way at the firm-wide level. They obviously would have recovered all the accumulated losses had they been kept. But they would be transferred as soon as the IG9 10yr skew would be anchored at zero, ie right on the “optimal gain” point for the plan of Dimon to produce the $60 billion of quite “tangible” gains in hard capital for the bank….The future concealments on the 10-Q VaR would corroborate that there was a plan in the “wind down” plan that aimed at maximizing the losses in the “tranche book” of CIO in quite a peculiar way: it paved the way of the future “London whale” scandal ahead of time.

a. the DTCC open position data shows that the IB was in front of CIO all along and adjusted its own positions to CIO changes in the course of 2012

The book was following two axis as per explicit repeated orders of Drew (20th January 2012), repeated again by Macris in early February 2012: “remove the short risk bias” and “cover the marginal default risk on HY”. As it will be shown down below, the original net “short risk bias” in the SCP would be covered in reality , not with selling protection on IG (even less on IG9), but with selling a total $50 billion protection on Itraxx Main ($19 billion on S9 forward spreads, $33 billion on Main S16 ($24 billion) and Main S17 ($9 billion)). This is a fact. This fact was known by the bank and it contradicts totally any future report done be that by the medias, by the bank or by any regulator involved. Since it was the very specific order of “Jamie”, the bank could have easily dismanteled the “IG9 led tale” of the “London Whale” conveyed initially by Rhule and Zuckerman. But the would never do that quite simple clarification.

The HY marginal default itself, that was 100% done to comply with Drew’s order again against the advice of Iksil, was covered using HY indices buying protection on $15-16 billion on on-the-run HY index equivalent ($8 done on HY17 itself,$7-8bln HY16-HY15-HY14). Iksil would warn many times still…. There was NOT enough liquidity on the on-the-run HY index of the time, ie HY 17, to execute the order of Drew given the target in time that she had given: “as quickly as possible”. She knew it well in January 2012 already. She would “listen” Iksil, “hear” Iksil and persist however… The resulting additional “short risk” that this US HY short was generating was balanced with $34 billion on the approved “IG9 forward spread trade” (June 2011), and MOSTLY $50 Bln in IG on the run indices (IG17 and IG18 US CDX indices).

Thus the long risk overlay was there in the book done in European markets, not US markets anyway…. And the HY protection as ordered was itself MOSTLY covered, NOT on IG9 despite the initial order of Drew, BUT on other IG indices. Here the reason was clear and known by Drew who had approved the switch from IG9 to on-the-run indices in the first half of March 2012: the IG9 index was notoriously NOT liquid.

This series of choices was NOT done at random and It would induce 100% of all the trades that would be done by the 3rd May 2012 on the “tranche book” of CIO. None of those trades would be unwound in relation to either the “freaking” moment of Drew on the 23rd March 2012 or the seminal press articles of April 6th 2012. That was somewhat “logical”. They all came from either “Jamie” or Drew and they had been repeated on and on since December 2011. The execution in practice had been achieved as per Artajo’s instruction day to day starting on December 9th 2011. Back then Artajo had explained that “Jamie” had turned “very bullish” with regards to the ECB LTRO operation. “Jamie” (Dimon) expected the European credit markets to recover very soon and very fast. The order of “Jamie” as per Artajo had then been: “this book must NOT lose money in the coming rally”. The choice to set the net long risk exposure on European Credit indices “Itraxx Main” was therefore the very direct result of “Jamie’s orders for early 2012”. But the book had remained slightly short risk in the US credit

79 index markets. That remaining short risk came from an order that Drew had stated on the same day a bit earlier, the 9th December 2011 in person in the central meeting room of the CIO London offices located at 100 Wood Street. She had said after the result on American Airlines: “Good! That’s what we want here at CIO…. Cheap options to gain on defaults… Renew it!” She knew that the liquidity was very poor since June 2010….

Thus on the whole the book had started the year 2012 being almost balanced in European markets and with mostly a net short of $4-5bln in HY indices. A good part ( but not all) of this net short in US indices, as ordered already by Drew since July 2011, was already mostly balanced by a long risk on “forward spread investment trades” (see Strategy 27 creation) on IG9 indices. Again that was done as per Drew’s former instructions. The outstanding amount in IG9 10yr index was ALREADY at $51 billion at the end of 2011 due to this specific order of Drew originating back from July 2011 (bigger than the total ICE/DTCC market size then!). Thus this “net short” in the US conveying already a huge exposure on IG9 10yr (as wanted by Drew since July 2011 in full knowledge of the liquidity issues) was balanced with a “long” in Europe as “Jamie” had wanted it as of January 1st 2012….

In order to avoid comparing pears and apples misleadingly for what comes, the description below will often refer to “CDX IG equivalent”. This means that all the exposures will be translated into an equivalent amount of risk as if the position had been held in “CDX IG“ indices. In that case of “CDX IG equivalent” exposures, it will be the current “on the run” index that will serve as a reference to compile the different positions. Likewise, in some other descriptions, the net figures will be expressed in “IG9 equivalent”, which will mean that the figures have been translated in a “risk neutral” fashion into amounts of CDX IG9 10yr index most of the time. Otherwise, it will be specified. For the experts, the translation used the ratios of risky durations, the ratios of spreads and the foreign exchange rate for European indices. It was made to be as simple as possible a proxy calculation. The rationale behind this formula was that credit spreads tend to move mostly in proportion day to day. This is why for example Jp Morgan and CIO would openly worry when CIO would breach what was called the CS10%W limit around the 22nd-23rd March 2012. That figure expressed the net exposure of CIO when one took this proportional move into account. That was another compelling sign that CIO was taking massive risk….2 good weeks before the very first seminal articles…

For each day that was considered, it was the credit spread of the day that was used. This is not a perfect calculation but it will be good enough for the descriptions that follow as the experts will see.

Let’s go back to the 1st of January 2012…..The book had then been still net short risk of about $20 billion (despite “Jamie’s” order of December 2011) in IG equivalent terms. That net short was decomposed as follows : $10 billion short in Itraxx Main, $30 billion short in CDX HY in CDX IG equivalent (or some $6 billion on HY index itself), and $20 billion long risk in CDX IG (all expressed in IG equivalent as explained above). The IG9 forward spread trade amounts to some $51 billion already (it will reach almost its apex size of $84 billion at end of February 2012… for the most part- ie not end of march 2012 but end of February2012 instead).

The main S9 forward spread trade is at $20 billion already. These already very large exposures had been ramped up mostly during the second half of 2011 as per Drew’s orders through Artajo instructions. If one wonders why “Jamie’s” order of December 2011 was not fully implemented before January 10th 2012 the reason would be known at once. Iksil had requested by the 15th December 2011 to wait until the 10th January 2012. Artajo had agreed. What was the “reason” given by Iksil to wait a bit longer to execute “Jamie’s” order? Iksil had pointed out that the biggest investment bank in Italy ( Banca Intesa or Monte Paschi?) was to launch in the markets a vital capital issue scheduled for the 9th of January 2012. If it failed Iksil feared that the rally expected by “Jamie” would never materialize. Indeed Italy would be perceived as being on the verge of financial bankruptcy

IF its biggest Investment Bank could not raise the capital that it had desperately needed. And actually Iksil had read in the press then that Jp morgan and Goldman Sachs had just withdrawn from the group of banks which had initially sponsored this crucial capital issue for this big Italian bank. Iksil argued that if ”Jamie” had doubts about this Italian capital issue as suggested by Jpmorgan’s recent withdrawal, it was inconsistent to expect a rally in Europe. Artajo agreed with the argument of Iksil and agreed to wait until the 10th of January 2012. Then indeed, one would know whether this crucial capital issuance was achieved or failed. The net short bias will be covered by Iksil as agreed between the 10th and the 13th of January 2012, using European credit indices….But soon after Drew will convey through Artajo the order to set the book longer risk outright….This long risk net bias will remain in the book based on the European credit indices, NOT the US credit indices anyway, in line with “Jamie’s” and Drew’s orders taken altogether.

What about the 30th March 2012 now?.....At the end of Q1 2012, the book is now long risk net $35 billion IG equivalent : $33billion long risk in Itraxx Main ($5 billion have moved out from tranche positions), about $110 short risk in CDX HY in IG equivalent (or $22billion in HY index itself), and $108 billion long risk in CDX ($88 billion added overall : 34$ added on the IG9 forwards, $50 billion added on the IG 16 and IG17 (the other $4 bln comes tranches delta changes probably). Thus one can see the move that had been operated on the book throughout the first quarter of 2012: the “long risk” added was done on Itraxx, ie the European market and NOT the IG9 or even other IG indices. The IG9 increase was “just” 40% of the total IG exposure that was actually covering the instructed increase in HY short. Why was the IG9 “only” 40% of the total? Because the IG9 index was NOT liquid….As Drew, Macris, Artajo, Hugues, Webster, Wilmot, Stephan, Kalimtgis, and others knew since March 2011 anyway (see again the story behind the strategy 27 creation)…

The table below describes how the positions grossly evolved through the first quarter. Please notice the already very elevated amount of the IG9 forward spreads on the left hand-side as of 31st December 2011. Numbers on the balances are expressed in IG9 10yr equivalent this time. They are all expressed in $Billions. (European notional are multiplied by 1.3 for the Forex effect between euro and USD. They are ALSO multiplied by 1.2 due to the wider spreads prevailing in Europe IG at the time as compared to US IG in order to provide ‘risk equivalent’ exposures)

Starting IG9 IG HY17 US Itraxx Itraxx balance Total point 10yr OTR eq balance Main S9 Main Itraxx in balance fwd IG9 10yr 10yr fwd OTR IG9 10yr IG9 10yr 28/12/2011 51 -30 -6 -10 20 -30 -10 -20 Amounts IG9 IG HY17 US Main S9 Main Net Total traded next 10yr OTR eq balance 10yr fwd OTR balance balance cumul YTD fwd &Xov Itraxx in 2012 er 17/1/2012 9.9 -1.8 -3.5 -13 6.2 9.5 15.7 2.7 31/01/2012 16 1 -5.1 -13 10 20.9 30.9 17.8 17/2/2012 18 20.2 -9.5 -17 13.4 19.9 33.3 19.6 29/2/2012 31 23.6 -12.3 -14.7 14.5 26.4 40.9 26.2 16/3/2012 31.5 25.7 -14 -21.4 18.9 20.1 39 17.6 30/3/2012 34 48 -15 -2 19 39.6 58.6 56.6

Against the legend that the bank conveyed through the media, the issue of sizes and risks was NOT at all reduced to an outright long risk on IG9 anyway, as far as the “tranche book” of CIO alone was concerned. A lot of quite equivalent amounts were present in tranches, in Itraxx Main, in Itraxx Xover, in CDX HY markets….Now if one considers the $3 trillion basis risk exposure of JpMorgan, one has to know that the only “reliable” skew quote in 2012 was the IG9 10yr skew quote. Thus the IG9 10yr really was just one position in

81 the “tranche book” of CIO among 4 others of the same magnitude at least, that was THE concern for the bank as a whole. That was an ongoing concern for all the regulators. The IG9 10 skew did not trade but many dealers spontaneously communicated “indicative” prices. This fact would be secured by CIO in March 2012 and Bloomberg chats or phone calls would prove it. That was the problem N01 for the market players and only indirectly related to 15% of the risks contained in the “tranche book” of CIO. Surely it would be the source of the largest losses in the “tranche book” of CIO however….

But let’s get back to the “tranche book” of CIO as a whole. The circled increases on IG9 10yr Fwd and “Main S9 10yr fwd” show that quite equivalent risks were present in other sectors than the IG9 itself. The Itraxx Main S9 index was more il-liquid than the IG9 and yet had been ALSO employed in order to execute Drew’s orders ‘as quickly as possible’. The root cause for the emergency was known eliptically: regulators “optics” and internal collapse (for Jamie’s share buybacks, being “priority No1” since March 2011) as Drew expressed it no later than the 20th January 2012. In the blue circled figures, one can see how this risk balance swung from 31st December 2011 till 31th March 2012: the long risk was added on “on the run” indices both on US CDX IG and European Itraxx Main IG indices due to a blatant want of liquidity on both CDX IG9 and Main S9 indices. Drew knew that issue since March 2011. And net-net, the long risk was MOSTLY added on European markets not US markets as per “Jamie’s” specific orders. One last thing to bear in mind is this amount of $51 billion that indicates that CIO had an exposure of $51 billion long risk on the IG9 10yr as of December 2011. As circled for the 29th February 2012, one will see that out of the $34 billion of IG9 10yr index that will be traded in whole over Q1 2012, $31 billion were done already by the end of February 2012. It matters for what follows in relation to the data that DTCC/ICE disclosed on the IG9 10yr outstanding exposure for the WHOLE market size. Iksil had asked to stop in full trading on the IG9 10yr after February 2012. Offering to be fired if needed, he would be listened to by a large extent this time. It thus shows how misleading the media reports, the bank reports and the regulators reports have been on this “IG9” matter all along. Iksil indeed had not been the sponsor of this IG9 10yr trade but instead the one who managed to stop it and this way before the first articles or the snowballing loss would appear. And Iksil here had quite material evidence to show that to just all the subsequent investigation teams. The latter would know of it way before 2014. They would all silence the facts until today in late 2017.

It is interesting now to contrast the volume traded by CIO versus the volumes cleared on those indices ( IG9 10yr and HY17 5yr for example) during that period. This gives an intuition as to what the market share of CIO alone was during the first quarter of 2012. The table below shows the amounts cleared on DTCC/ICE and the volumes traded by CIO for the CDX IG and HY indices. The data on clearing were disclosed on a weekly basis. There was thus a delay between the actual data being made available and the real timing of the related trades. It matters for some quite critical days, especially the 29th February 2012 which appears to be a turnaround in the actual manipulation that occurred in the markets. For example, the volumes traded on the 29th of February were cleared on the day labeled as “March 6th 2012”. This is quite a peculiar coincidence when one remembers that the 29th February was the day of the “CIO Business Review” chaired by Dimon, presented by Drew. The 29th February 2012 was also the day when CIO traded unusual volumes on IG9 10 yrs BECAUSE Boaz Weinstein quite visibly tried to push the prices of the IG9 10yr right before the New York close for month end. This suspicion against Weinstein is NOT solely the conviction shared by many among the CIO staff at the time (see Grout, Artajo, De Sangues on the matter outside of Iksil). This is the day indeed when a Weinstein’s long term alumnus felt so bothered by what his mentor had done that he (Gabriel Roberts head of index trading at CITIGROUP then) felt the duty to call Iksil with a senior salesperson (Karim Lahcene on March 1st 2012) to tell him that “only one guy” was lifting prices at this NY close then. Roberts specified on the record then with Karim Lahcene (CITIGROUP) to hear it, that he Roberts did NOT understand what Weinstein was doing here. Roberts specified also on the record that, on the contrary, he perfectly understood the course of action of Iksil then. This anecdote that has been systematically silenced by the bank and every single regulator involved since

82 then disproves totally the allegations of “potential market distortions” that the CIO trades might have caused in a quite hypothetical theory (see the “settlements” of October 2013 involving the bank, the FCA and the CFTC). Yet the transcript of this phone call of Roberts& Lahcene to Iksil has been seen by all the investigation teams.

Now the day of March 6th 2012 is THE day when Drew sent this instruction AGAIN to alter the estimate P&L process to Artajo. Yes, that was the second time Artajo conveyed his orders to Grout and one wonders why really. The first time (likely around March 5th 2012) it was Grout who had been the direct recipient of this order from Artajo. That was plain natural since it concerned Grout’s job at CIO (not Iksil’s job) and Artajo was Grout’s direct boss in effect (against what the org chart of the bank pretends). That second time, as of March 6th 2012, Artajo was putting Iksil in the loop. One must wonder why…. Artajo would call Iksil for him to convey this order to Grout a second time ie the very same orders while this was NOT Iksil’s duty, while Grout had already understood HIS duty here and while Iksil was actually on vacation 1000 km away thus mostly unaware of what was going on at CIO. One can picture the scenery… Grout and Artajo sat 10 meters away in the open space of the CIO London office. Drew was quite active in preparing to surrender the “tranche book” to Ashley Bacon the deputy firm-wide CRO. Grout and Artajo spoke several times a day then to address Drew’s urgent follow-up requests. And at one point in time, not the 5th but the 6th March 2012 in the afternoon, Artajo called Iksil from his mobile phone. Iksil had been for 3 days in a ski resort in France, 1000 km away from both men. Here Artajo called Iksil, informing him summarily of what was going on in New York and next instructing Iksil to call Grout on the usual line and REPEAT the recent Artajo’s orders to Grout a second time. Artajo invoked then to Iksil the fact that maybe Grout had not fully understood his boss Artajo due to “language barriers”. As Artajo had told Iksil, this call was just meant to secure the understanding of Grout over the instructions of Artajo’s from the 5th of March 2012. Iksil could not and would not check whether Artajo was truthful here. Iksil would comply with Artajo’s order, call Grout and convey the instructions of Artajo once again.

The record of the call shows that it took one second for Grout to assure Iksil that he had plainly understood this order here about the price selection process that had changed the day before actually. Still the bank and just all the authorities involved will allege that it was Iksil who had conveyed this instruction to Grout and it was therefore Iksil who had a “responsibility under his management for implementation” in that regard…..How honest were they all in this allegation here after reading the transcript of this call in full? And this call of March 6th 2012 will be scrutinized a lot as early as late 2012, ie before the Task Force Report (January 2013), before the US Senate Report (March 2013), before the august 2013 charges, before the “settlements” of the bank in September-October 2013 shall be published….As of late 2017, the truth is not told on what was a key evidence for just all the investigation teams. The call was “transcript-ed” accurately in general.

Drew would always deny having conveyed such an instruction to alter the price selection process on the 5th or/and the 6th March 2012. But the March 6th 2012 day is ALSO the last valuation day that Drew would use on March 20th 2012 for HER DRPC presentation to the board. What a coincidence for someone who always denied her knowledge of this change that was effective on March 6th 2012 precisely so. She did choose that date in particular for reporting valuations to the board of Jp Morgan though. What is strange here also in her own choice is that the US Senate report mentioned that she would use figures dating from 2011 year end with regards to the other risk figures…… What a mix Drew did here, randomly picking March 6th 2012 for the valuation and December 2011 for just all the other risk figures, while she had all the numbers for a more recent and consistent date like say the 13th March 2012 or even later…. This March 6th 2012 day, as said, was ALSO the day when ICE/DTCC would publish its clearing data related to February 2012 month end when Weinstein apparently attempted to manipulate the closing prices of New York. The current net market share of the bank on ICE/DTCC may have mattered to the board members then in relation to the “’tranche book” of CIO. That was the 20th March 2012. Therefore, the amounts provided by DTCC/ICE aggregate numbers happening up to one

week later should be reset accordingly in particular for that critical week spreading from February 29th till March 6th 2012:

Period dates IG9 ICE IG17 ICE HY17 ICE CIO CIO mrket start: Jan 1st 10yr amounts 5yr IG17 traded HY17 market share on IG17 traded cleared traded Amounts by CIO cleared share on and HY117 by in total by CIO cleared in total IG9 CIO in total 17/1/2012 9.9 65 -1.8 88 -3.5 15.5 15% 12% 31/01/2012 6.1 75 2.8 104 -1.6 13.9 8% 7% 17/2/2012 2 67 19.7 139 -7.9 16 3% 32% 29/2/2012 13 34 3.4 75 -2.8 20 38% 9% 16/3/2012 0.5 33 1.9 76 -1.7 16.6 1.5% 7%% 30/3/2012 2.5 19 22.3 106 -1 22.7 13% 11% Total Q1 34 293 48 588 -15 104.7 8.6% 11%

The month of February 2012 now is ALSO worth a serious look at. It is the month when Drew, Macris and allegedly Dimon wanted Iksil to trade on and on despite Iksil’s repeated alerts and suggestions to “stop trading”, “take the full pain” in loss terms now, and “let the book die” for good in this “liquidity trap”. Artajo would be secretly demoted…. Despite all this “push” inside Jp Morgan from all ranks to trade “IG9” again and again, CIO happens to be “only” 25%-30% of the market activity from mid February till end of February 2012, yet not all the time (here the figure for the 17/2/2012 is compounded with the one for the 29/2/2012. To be sure the public data from ICE/DTCC came out a week after). On average, CIO was about 10% of the market trading volumes. These figures of market share above are representative as CIO mechanically used ICE for its index trades or DTCC in 2012. The figures above show, as Iksil testified, that at FIRST the “buy HY protection balanced in risk terms with IG indices” combined order of Drew since early January 2012, is here executed NOT on IG9 but mostly in “on the run” indices. NEXT, around mid February 2012 as per express order of Artajo, the IG17 part is switched thus from “on the run” to IG9 10yr index mostly. Against the legend again that pictured CIO as trading like mad, one can notice that the CIO market share collapses in the first half of March. The “seminal” articles structurally conveyed a fallacy in that CIO had already stopped for long trading in size on the IG9 in mid March 2012. Thus, at the peak of its activity on IG9 10yr, CIO was only a third of the market volumes and this occurred for one week in late February 2012 for the whole first quarter of 2012. And this spike is in fact totally due to the quite aggressive behavior of Weinstein towards the NY close on the 29th February 2012 that even his long term fan Gabriel Roberts could NOT explain.

The Totals at the bottom prove that any pressure on price from CIO would have been very temporary if there was any. The pressures, if any, would have solved by the mid March 2012 in any event. This also disproves the mischaracterizations of the Bank, of the FCA and of the CFTC in October 2013, however ambiguous and suggestive they were in their public statements.

The only remaining issue would be that the size traded at February Month end was allegedly meant to flatter the P&L. This requires a finer analysis that will be done later in this document. Yet, as such the volumes traded by CIO during Q1 2012 were not extraordinarily high given the turnover on the IG9 10yr : CIO weighed 9% to 11% on IG9, IG17 and HY17 over Q1 2012. This disproves that CIO created a distortion on the sizes traded: the turnover is 10 times higher during the quarter. Yet, against the accepted assessment that the IG9 was “not illiquid because allegedly it was trading for “$10 billion per day”, one can see here that this assertion alone is plain wrong. No, the IG9 index could certainly NOT be “deemed” liquid. The bank knew it. The regulators involved all knew it. And yet they will all carve in their reports of April-May 2012 that the IG9 was “still deemed liquid”. Doing so they fueled this artificial mismarking story that will cause the no less

84 artificial restatement of July-August 2012. They all used the “$10 billion turnover per day” argument to justify their characterization. But the data here proves not only that they were wrong but ALSO that they knew they were wrong. The IG9 and the Itraxx Main S9 and almost all the HY index positions were il- liquid in 2012 and therefore mandated a $4-5 billion liquidity reserve for the “Tranche book” of CIO.

For the sake of completeness -for this crucial point as it shows the genuine mismarking that would hidden behind the false mismarking- here is the OCC report that was dispatched through the regulators chiefs as of April 17th 2012. The extract (can be retrieved among the exhibits of the US Senate Report) below describes what the OCC examiners would agree with Drew to tell in the future. In Red and Bold are the wrong statements. In Bold and Black one will read the right and key statements. The OCC had met with “management at CIO” on April 16th 2012. As one will see: Iksil was NOT part of “management” despite the fake Org-chart that the bank would set up in late 2011 : “4/19/2012 5:53:56 PM From: Crumlish, Fred Sent To: ; Subject: FW: JPM CIO IG9 "whale" Trade Doug FYI. I thought I did copy you on this. Ron - just if you're curious. Prop or not prop, that is the question ...

Tuesday, April 17, 2012 04:33 PM From: Crumlish, Fred sent To: Brosnan, Mike; Belshaw, Salty; Pfinsgraff, Martin; Waterhouse, scott Cc: Wilhelm, Kurt: Banks, George; Fursa, Thomas; Hohlt James; Kamath, Jairam; Kirk, Mike; Monroe, Christopher; Swank, Todd; Wong, Elwyn Subject: JPM CIO IG9 "whale" trade On Monday 4/16 OCC and FRS examiners met with Ina Drew and several members of CIO staff and risk management to discuss the JPM synthetic credit book in view of recent press reporting. This message provides a summary of our discussion, followed by a more the detailed summary. It focuses specifically on recent changes to the synthetic credit book. JPM's CIO has been using a synthetic credit (credit derivative) portfolio since 2007. It was initially set up to provide income to mitigate other significant credit losses that would surface under a broad credit stress scenario, Since it wasn't possible to tailor a specific hedge to the JPM balance sheet as the time this portfolio was constructed. As the investment portfolio grew in 2007-2009, the synthetic credit portfolio was used to hedge stress and jump to default exposures in that portfolio as well. ClO's credit derivative position was managed to provide around $1 billion to $1.5 billion income in credit stress scenarios against firm wide losses of $5 billion to $8 billion, In late 2011, in view of a change in perception in the state of the economy, CIO managers decided to reduce high-yield (Hy) credit protection; however, after the AMR bankruptcy and the Kodak expected to file for bankruptcy, the markets for CIO's HY indices weren't Iiquid enough to use them to unwind that position. The IG 9 index, which is much more Iiquid than HY indices, includes five 'fallen angels" that allowed it to be used to reduce a "good part" of CIO's HY position, so it was used to reduce the HY protection. The IG 9 market is not il-liquid as it trades around $10 billion daily and spread changes for this index are in line with peer indices. The IG 9 curve has steepened in a move of around 6.5 standard deviations, and there has been strong buying of deferred contracts, implying that the buyers are certain that there will be no defaults in the next 9 months and nearly certain that there will be defaults next year. In view of events, however, JPM is conducting a "post mortem” of the IG 9 situation and its impact and share results with OCC and when completed.

Comment: the very wrong statement that IG9 was “still deemed liquid” and was trading $10 billion a day is here highlighted in red bold. The OCC knew that it was plain wrong if only because ICE data indicated a much, much lower level of turnover. The descriptions that follow on the lines of “spread changes for this index are in line with peer indices” prove that the allegation is knowingly false actually: a “6.5 standard deviation move” plus the very weird certainties of market players about defaults show that this situation is SPECIFIC to the IG9

85 and does not apply to any other existing index. Therefore, “No”, this index was NOT in line with peer indices at all. These are as many intuitive signs that the IG9 10yr was il-liquid. As a very direct consequence of that ALSO “this index” IG9 was NOT “in line with peer indices” that were “liquid” indeed. None of this was new. Iksil had warned since January 10th 2012 on the matter. This is why Iksil finally succeeded in pushing the CIO management to start an official post-mortem of the “IG9 situation”. That “post mortem” was started no later than the 26th March 2012 right after Drew had ordered to “put the phones down” on Friday March 23rd 2012. As they met on April 16th 2012 the OCC and the CIO chiefs were thus in their 4th consecutive week of Post-Mortem precisely because the IG9 10yr index and additional 60% of the “tranche book” positions were il-liquid.

The April 17th 2012 summary of the OCC “knowledge base” goes on:

‘The CIO began using credit derivatives around 2007 as part of its mandate to manage structural balance sheet positions. CIO partly uses credit derivatives on indices, not specific names. Initially CIO bought protection (shorted risk) on mortgages, using ABX, and high yield indices to mitigate some of the firm's balance sheet credit exposure. At this time CIO investments were highly concentrated in Agency pass-through mortgage securities, and the structural credit risk was in the lines of business. Through the financial crisis deposit inflows combined with lower loan demand to leave the firm with significant excess funds. As part of its mandate to invest, when appropriate, in high credit quality, liquid investments, the CIO began purchasing low credit risk, top of the capital structure securities to use the excess funds, although of high quality, these investment securities have more credit risk than the US Agency pass- throughs that continued to be held, so that structural credit risk in the investment portfolio increased along with portfolio growth.

Other key comment: the OCC had the inventory of the investments of the CIO here. The OCC vastly understates the fact that those new investments were NOT liquid notoriously so since the crisis of 2008. More these il-liquid investments of CIO were even more concentrated. Those investments were often CLO tranches of CMBS or illiquid RMBS where CIO often held 80% or more of the whole issue each time! Thus CIO was quite visible, highly concentrated on AAA tranches that were known to be structurally il-liquid.

The April 17th 2012 summary of the OCC “knowledge base” goes on:

“Throughout this the CIO continued using index credit default swaps (CDSs) to mitigate some of the structural credit risk in the investment portfolio and the lines of business other than the investment bank, which manages its own credit risk exposure. Although there are liquid markets for many credit derivative indices, the markets are not deep enough to fulIy hedge a multi-trillion dollar balance sheet. CIO's credit derivative position was managed to provide around $1 billion to $1.5 billion income in credit stress scenarios against firmwide losses of $5 billion to $8 billion. CIO managers decided to reduce the high yield credit derivative protection around Thanksgiving last year. After the AMR bankruptcy filing on November 29,2011, the firm profited from its credit derivative positions as anticipated, but high yield index derivatives had limited liquidity as demand increased, CIO managers thought that it wouldn't be possible to reduce the high yield credit derivative position by using the indices that created it; the best available hedge product was the IG 9 index, which has good liquidity as an investment grade index and a high yield component as five of the index companies are "fallen angels" i.e., companies that have fallen below investment grade since the index originated.”

Last comment on this report: The statement in red right above is again very wrong and the former description of facts related to how the book was morphed as per “Jamie” and Drew orders proves it once again. No the IG9 was not having “good liquidity” at all. The CIO chiefs mischaracterized the facts and the OCC was keen to gobble the tale. It was NOT any longer the “best available hedge product” since late 2010. Iksil had warned many times about it no later than March 2011. The strategy 27 would be created inside CIO by July 2011 precisely to prepare the post-mortem of the “tranche book”. The IG9 legacy positions could not be unwound in the markets. And Iksil would repeat it again by the 9th of December 2011 in front of Drew, Macris, Wilmot, Artajo and others face to face. Then as it was already impossible to unwind IG9 positions and “credit Hybrids” was NOT marking its own IG9 positions where other dealers were quoting it, the picture was obvious. The evidence of that exists among the Bloomberg chats involving Iksil with IB traders in writing. In the former

86 descriptions in this document it was indicated first that the Itraxx S9 index ALSO was used although being even LESS liquid than the IG9. Why was that? Because the IG9 itself was really well short of the liquidity that CIO needed. Thus any other less liquid index like the Itraxx Main S9 would help… Any little help was welcome outside of the il-liquid IG9 index market. Second, the January 11th and 12th 2012 Bloomberg written chats with David Goldenberg would show that the ongoing activity on IG9 was close to nil already.

At last, the OCC here recognized a very important knowledge of a situation that dated back from 2007: the CIO “tranche book” could NEVER be large enough to effectively hedge the systemic risk of the bank. And here one can see that the OCC hints at the “basis risk”, or the $3 trillion notional amounts of credit derivatives that are NOT necessarily fungible. The simple sentence of the OCC indicates that the US regulator was very close to the “wind down” plan of Dimon that targeted the “tranche book” of CIO since September 2010. This is not a benign remark at all. This point here matters a lot and the reader should look once at the “Var History” document on this website to understand the magnitude of what this sentence conveys. In short this simply means that both the CIO and its “tranche book” initiative as it started in 2006 was as much the initiative of Dimon as the one of the regulators especially of the OCC. This sentence here provides a basis for understanding also why the “post implementation review of the NBIA” has never been even started by anyone at JP Morgan. Thus in mid-April 2012 all this complacent pretence that the IG9 index was “still liquid” is here to hide a longstanding failure of the “hedging initiative of JpMorgan Balance sheet systemic risks”. This failure was known since 2007: “the markets are not deep enough to fulIy hedge a multi-trillion dollar balance sheet” The “tranche book” of CIO, once built-up in mid 2007, was known to be itself il-liquid and therefore pointless at the very best with regards to the initial strategic project. It was therefore suspected since then to be just a “profit maximisation” tool, ie since 2007….Why did it take 4 to 5 years for regulators to wake up officially?

There is one simple answer: there was no clean solution. With regards to this strategic liquidity issue here, things had just got worse since 2007 actually. In early 2012 even the on-the run indices were “’barely still liquid”. The data shows it as well. First one can see that the most “liquid” IG17 index had a total turnover of only $588 billion for the first quarter, ie for some 60 active days. This gives an average traded volume of only $9.5 billion. The OCC and JpMorgan top managers said it above: a credit index was “deemed liquid” if it could trade $10 billion a day at least….This means that the IG17 itself was not “deemed il-liquid” but just by a really narrow margin. As to the IG9 10yr, DESPITE the alleged flurry of activity of the CIO, its average was twice smaller to $4.8 billion. Yes the IG9 10yr was “il-liquid”, no doubt. Without the CIO allegedly “crazy” volumes (ie $34 billion traded on the first quarter of 2012) the average would have been $4.3 billion, not a big sea change anyway…. Clearly all the regulators saw all that every single day of 2012 through the ICE/DTCC publicly disclosed figures: the IG9 10yr was twice less traded than the “barely liquid” IG17 was. The IG17 was deemed as “not il-liquid yet” for regulators and the bank as per their self chosen criteria….fair enough… Let’s not be mean right? $9.5 billion a day is “almost” $10 billion a day….But the IG9 10yr was clearly il-liquid and the figure of $10 billion of traded volumes per day was plain wrong as attributed like this to the IG9. They all could easily check on it in the second quarter of 2012…They did and conveyed a very misleading picture here, be that Drew, JpMorgan and the OCC altogether. Not a single authority or other regulator will clarify that misrepresentation here.

The next chart displays the history of the cleared volumes on IG9 10yr. One will notice a first spike in the middle of January that cannot be explained in its magnitude by the sole activity of CIO. The chart indeed below displays the amounts cleared in weekly totals and the remaining outstanding exposures in the markets after all nettings had been processed between the registered participants. CIO was one of them surely so. These figures, even though they did not represent the whole market, were representative of say 90% of the market activity in Q1 2012. Thus in the course of January 2012, CIO traded say $10 billion of IG9 10yr. The chart below shows that the amounts “cleared” totaled about $50 billion for this month. This is the period when the VaR of Jp Morgan blows up irrespective of the changes occurring in the sole “tranche book” of CIO. This is when supposedly the “tranche” part of “credit hybrids” is “off-shored” (the chosen term for the FCA-

“externalization” for Macris, Pinto, Hogan, Bacon or Artajo- “wind down” for Drew, Bacon and Dimon) to hedge funds like Blue Mountain. Thus one can infer here that maybe up to $40 billion of IG9 10yr index were “cleared” on behalf not of CIO at Jp Morgan but on behalf of the defunct “tranche” business of “credit hybrids” towards hedge funds like blue mountain or SABA (Weinstein’s hedge fund). If true that “balance” suggests that JpMorgan was involved in most of the IG9 trades in January 2012. That alone was enough to set on red alert all the market players. Even if then the IB had “only” traded $10 billion in front of the $10 billion forcefully ordered by Drew then, the “Jpmorgan vs JpMorgan” marionette show occupied $20 billion out of $50 billion traded volume, ie 40% of the activity. That was really way enough to drag the attention of all the “trend followers” like Weinstein or Vilani. That was enough to catch the focus of Hubbard, and others heavily exposed to the IG9 10yr skew already. Stephanie Rhule then, getting information from Weinstein to Hubbard, was monitoring the tale in the making already too.

There was a lot to see in January 2012. One has to notice ALSO that at the start of 2012, CIO has $51 billion in IG9 10yr while the total market size is only of $35 billion. Is CIO larger than the planet on IG9 10 yr? No, this is plain impossible. Just maths: no-one in the market can be bigger than the whole market, not even JpMOrgan itself…. Those 2 figures, $51 billion for CIO alone and $35 billion for the whole market, simply tell that Jp Morgan as an entity was much, much smaller on IG9 10yr than CIO alone was. The IB certainly had offsetting exposures on the IG9 10yr that had a magnitude similar to the CIO’s, if not bigger actually. One should assume here that no regulator would have let Jp Morgan be larger than 33% of the whole market. This means that Jp Morgan had a net total exposure on the IG9 10yr that would not go much beyond say $11 billion or so. Thus the firm had another $40 billion exposure, away from CIO and going OPPOSITE to CIO, on the IG9 10yr at the very start of the year 2012. This situation was a legacy of 2011. The “off-shoring” of the defunct “tranche” business of the “Credit Hybrids” book would not change the market net figures but it may have changed the next exposure of Jp Morgan as such. In any event again, ICE scrutinized that in its role of “clearing agent” for all and every regulator involved would worry and monitor this quite closely. Here is the chart:

13/12/2011 12/01/2012 11/02/2012 12/03/2012 11/04/2012 11/05/2012 10/06/2012 10/07/2012 25 000 000 000 80 000 000 000 IG9 10yr volume cleared 75 000 000 000 IG9 10yr Oustanding 20 000 000 000 70 000 000 000 65 000 000 000 15 000 000 000 60 000 000 000 55 000 000 000 10 000 000 000 50 000 000 000 45 000 000 000 5 000 000 000 40 000 000 000 35 000 000 000

- 30 000 000 000

13/12/2011 27/12/2011 10/01/2012 24/01/2012 07/02/2012 21/02/2012 06/03/2012 20/03/2012 03/04/2012 17/04/2012 01/05/2012 15/05/2012 29/05/2012 12/06/2012 26/06/2012 10/07/2012

On can see that the “cleared” amounts are given with the left hand-side scale, and the outstanding market size appear on the right hand-side scale. The first spike of activity occurred in January 2012. Next there is a spike on March 6th 2012 that, as explained, pictures what had happened on the 29th February 2012 in fact. Still one can notice that the next spike of clearing volumes occurred in the second half of March 2012 actually, right when CIO was NOT trading IG9 much any longer. Some more trading occurred for some size AFTER CIO had stopped trading actively. That spike was occurring in the first 2 weeks of March 2012 once one has considered the 1 week offset. Thus in both cases of activity spikes, the first one in January 2012 is too big to be attributed to CIO alone, and the second one occurred right when CIO had almost stopped trading totally. Indeed for Mid-to- late March 2012 the increase in cleared volumes was NOT happening when CIO was trading. Surprisingly so, one will see above that throughout the media campaign the turnover was relatively low (April and May 2012) although the losses ballooned right then in CIO. Thus the market was NOT trading much and prices changed a LOT…. That feels very much like a targeting first, a manipulation next on prices for a well anticipated capitulation…

And what happened after March actually? Well the “seminal” articles of Stephanie Rhule and Greg Zuckerman would do part of the job. It remains that the market size barely changes over the period. This indicates that no- one took profit or grew the position while the prices were being moved much more than in the past. This is another evidence of market manipulation where clearly the CIO was completely absent other than in the press reports. This is strongly indicative of a rigged market whereby few players trade one with each other just to push prices in limited volumes. David Goldenberg (CSFB) on January 11th 2012 seemed to predict that outcome if one can read his Bloomberg chats with Iksil. Andy Hubbard, head of index trading at CSFB, Goldenberg’s boss, and husband of Stephanie Rhule, would therefore certainly explain clearly how such a price drift on no volume occurred in late March and in April 2012.

One can see as further evidence of that with the steep decline in the market size in the first half of June right when the market “heard” that Jp Morgan had “got rid” of say 70-80% of the positions. Who in the market ignored that this was just a “JpMorgan vs JpMorgan marionette show” here? That certainly was a “mistake” to run the show like this.

As reported in an article mentioned formerly on the website, and in this document as well, the traded volumes were notoriously quite lower than the $10bln figure adopted by the bank and the regulators so officially. One can see by the way on the chart above how wrong this “$10 billion per day” assessment was about the IG9 10yr: most of the time the weekly cleared amount ( ie for 5 sessions in total) does not go beyond $10 billion actually. How could it be then that the IG9 10yr trade volume could ever be $10 billion per day?! The ultimate spike of cleared amounts occurred right when “officially” the CIO “tranche” positions had been moved to the IB and to hedge funds like Blue Mountain on the follow sometime in June 2012. Feldstein from Blue Mountain would soon publicly state that here he had just been a “facilitator” on the matter. That really was just a marionette show titled “Jp Morgan vs Jp Morgan”… This is when the market size went back down to billion and when the speculation against CIO was over for good. All this data shows a very different story than the legend that both the media, the regulators and the bank conveyed up until 2017.

No, the IG9 10 yr was NOT liquid at all (even the benchmark on-the-run index was not liquid even though it was not il-liquid for sure). No, the CIO did NOT influence the market volumes much and even less the market prices. But, JpMorgan, Weinstein and Dimon strategy did have a material moving effect. No, the CIO did not make the market size grow or decline throughout the first 2 quarters of 2012. The mentions of January and mid Marhc 2012 show it among other things that have been described around Weinstein, Hubbard, Goldenberg and Rhule activities. The market size increased and declined in direct relation to the long planned internal “off- shoring and wind-down” operation. This would be ultimately a collapse engineered between CIO and “credit

89 hybrids” on “tranche related” positions that had been present inside Jp Morgan for years already. No the regulators were not unaware: they were active participants in this plan since 2010 at the latest. Instead they monitored the data and scrutinized the liquidity. Yes, they knew that their shared assessment with the bank that the IG9 10yr index was “still deemed liquid” was plain wrong. They saw as well that doing so they knowingly ignored a mismarking on liquidity reserves that pervaded throughout the bank since 2007. They knew it all along in 2012 and knew it again when they ”settled” with the bank in October 2013. They thus may NOT have wanted to know how clear it was that CIO had NOT traded in order to favor its estimate P&L for “good reasons”. Yet the evidence was there right under their eyes for years now.

The following table clears the allegation that CIO traded at the end of February 2012 in a way that flattered its estimate P&L. Some written chats and the call of Roberts dated March 1st 2012 will corroborate that. Once it is acknowledged here that if the IG 17 volumes were so low that it was NOT really liquid and that the IG9 10yr in fact only traded at best $2-3 billion a day without Weinstein’s suspicious activism, one realizes that simple fact that the IG9 10yr was NOT liquid at all. As Grout would put it: “it was a no man’s land”… almost that indeed... The weeks having more than $10 billion of traded volumes would be scarcer and scarcer…Yes, “weeks” , certainly not “days”…

The table that follows thus puts the figures in context as to whether the flurry of trading of CIO in late February 2012 did change the picture for CIO or not in terms of performance. Here the “performance” for the IG9 is expressed in Bp (0.01%): a positive number means “outperformance” for CIO while a negative number indicates a loss for CIO. 1Bp of underperformance in the IG9 10yr position alone was worth a loss of $45-50 million for CIO by the end of February 2012. 1 Bp of loss on the so called “IG9 forward spread” was worth a $25-30 million loss for CIO (the “IG9 forward spread” included the IG9 10 yr position along with other IG9 balancing positions. This included the IG9 7yr, the IG9 5yr and IG9 tranche positions. The ratios had been scrutinized and determined ultimately in June 2011 by Stephan, Kalimtgis and Artajo on behalf of Drew and Macris. These ratios would be adjusted next by the same persons, not Iksil who was there to execute as per those ratios- (Iksil had had to suggest ratios initially back in March 2011 that would NOT be endorsed by CIO next).

ICE ‘batch’ Cleared Outstanding IG9 Fwd CIO trades Short risk Cumulated dates (from Volume in in Bln for apparent in IG9 10yr guys alone IG9 “fwd 11Jan2012) Bln IG9 10yr performance (best guess) spread” vs market performance peers 1/12/2011 35 0 0 ? 0 Year end 35 +2* 0 ? +2 6/01/2012 0.5 35 +3** 2.3 bln +5 11/01/2012 11 37.3 +2 6.5 bln +7 18/01/2012 14 41.2 -4 1bln -3bln +3 25/01/2012 10 42.9 0 2.2Bln +3 01/02/2012 20 50 -1 4.4 Bln -2.6bln +2 08/02/2012 6.3 52.2 -5 0.6Bln -1.6bln -3 15/02/2012 6 54.6 +3 0.4 Bln -2 bln 0 22/02/2012 9 56.3 -0.5 2 bln 0 -0.5 29/02/2012 7.5*** 60.9 +1 11.5 bln -4.5 plus? +0.5 7/03/2012 15.9 70.8 -4**** 0 -5bln -3 14/03/2012 8 69.8 +0 0.2 0.8 -3 21/3/2012 11 68.6 -4 2.5 ?!? -7 28/03/2012 14.7 70.7 -2 0.5 -1.5 -9 03/04/2012 5 71 -1 0 0.3 -10  * this is natural noise into year end. Favorable to IG9 as it was the trend since August 2011 actually.

 **The 3bps here come from the HY squeeze.  *** 29th February : the last $4.6 billion of SABA and the other 1.2bln traded after 4.30PM London time ( total 5.7 bln) probably appeared on the next batch occurring on March 9th 2012. Thus only 5.8bln of CIO went into the batch including the trades done on the 29th of February.  ****Manipulated or not? It seems that Grout should have put a further 3bp hit in normal times. Please notice that the IG9 underperformance “versus peers” is more than 7bps since January 12th 2012. This is quite abnormal.

The table above lines up what “Iksil” saw in his day to day job talking to Grout with the column labeled as the IG9 10yr index performance versus its peers. It also displays what Iksil kept elevating with the right hand-side column referring to the “IG9 forward spread” drift versus “peers” again. The figures above are just recollections. But they should be quite reliable as Iksil worked on the issue every day since the very first week of January 2012. This analysis motivated his alerts and was the most important part of his job actually. As it was mentioned, irrespective of the CIO alleged “flurry of trading activity on IG9 10yr”, the 5-day traded average did not go much above a total of $10 billion per week. It is never enough to repeat it: the IG9 10yr index was NOT liquid at all in early 2012. The fact was notorious. The firm knew it since 2010 and so did the regulators (see the “off the run” rule enacted by Cavannagh across Jp Morgan by March 2010). The index was quoted in general in a 3Bps “bid-offer” for a trade size of $300 million since early 2012. This followed a long period between August 2011 and December 2011 where trading on the IG9 10yr was almost impossible. The quotes in 2012 were mostly “indicative only”, or “unprintable” from the Bloomberg terminal, ie “unreliable” for straight execution anyway or “immediate valuation”. Not many trades occurred in the course of the day to be sure in 2012... The intrinsic uncertainty on the price was at least the “bid-offer” since very little volume would trade on one “bid-offer” quote anyway. Thus when one reads the middle column displaying the performance of the IG9 10 yr index versus its market peers, one sees that there was really little fluctuations (between +3 and -5) overall. One can see that it looks like a random walk over the whole quarter. Yet the CIO “tranche” book lost money quite regularly over the quarter, week after week…..

What was happening? That was a sneaky “drift” that could be “felt” but that also was quite hard to single out on a daily basis in price terms. That left Grout regularly puzzled and undecided as to which price to pick at the end of the day. That consequently left Iksil in the blind all day since Grout could not provide intermediate pictures of what the performance of the strategies was in the course of the day. Iksil waited for Grout to assess the “trades to do”. And Grout waited for Iksil trading experience to refine his own analysis on price and performance. Both were in a deadlock in 2012 each day: there was no liquidity, and therefore no reliability in the quotes that they received. What to do then? Well, Iksil elevated soon enough as of January 20th 2012 that the “IG9 forward spread” was actually losing ground in quite a sneaky way. Here this was a loss that was quite specific to CIO and was based upon IG9 5yr and IG9 7yr indices that, unlike the IG9 10yr were actually outperforming their peers quite regularly but almost without trading. That felt already like a manipulation targeting CIO positions.

The unsolicited warnings coming from Goldenberg (CSFB) dated January 11th 2012 left little room for the doubt. A manipulative intent targeting CIO was under way. It may not succeed as incidents of this kind had occurred on 2009 and had failed. It would be tough for CSFB and others to succeed unless the IB of JpMorgan was targeting CIO too. The markets were il-liquid for everyone indeed. They could only succeed by moving the consensus prices in synch among the main dealers without trading. They had to have the IB of JP Moran on their side on this matter of consensus prices being moved like this. That was tricky to hit the CIO specific positions as they involved many instruments all at the same time. No matter indeed how il-liquid the IG9 10yr was, the IG9 5yr and IG9 7yr were anyway much more il-liquid than the IG9 10yr. The IG9 outperformed the market “peers” on the ‘7yr ‘ and ‘5yr’ tenors but underperformed on the ‘10yr’ tenor….

Surprise, in January 2012 (unlike January 2009), the dealers seemed to be successful….The combination of the il-liquid IG9 10yr prices and the “much more il-liquid” IG9 7yr and IG9 5yr, gave way to a very regular “non random at all” walk on the “IG9 forward spread” that was THE position of CIO. That was a sure loss to show in the “tranche book” of CIO. Everyone got it as of January 20th 2012 when Iksil explained in a conference call involving Drew, Macris, Artajo, Stephan, Hagan, Goldman, Kalimtgis and others….In full knowledge then of the high trading cost that would induce even more losses in the “tranche book”, Artajo will order Iksil to keep trading in order to “prevent the IB from moving the prices against CIO”. “That was the IB. Drew was arms twisted internally to settle with the IB and surrender the book. CSFB and others were just seeing their interest here.”….So was the theory of Artajo at the time.

Drew had a different perspective than Artajo. But one that related to the very same issue. She will instruct Iksil to proceed on HY and the “long risk” at the same date in order to match with the “regulators optics” or regulators’ “expectations”. Macris will order Iksil to keep trading as well being mostly concerned by the snowballing losses….He Macris had another perspective on the very same issue, being obsessed with the loss. For all of them Iksil had to keep trading despite all his alerts in a context that was very well understood by all the decision makers: regulators’ optics, “wind down” plan, IB and losses at CIO. They surely all connected the dots here.

One can contrast the “random walk” performance of the IG9 10yr (column in the middle of the table) versus its market peers and the quite regular underperformance pattern of the related “IG9 forward spread” versus the market peers as well. Overall one may summarize Q1 2012 saying: “well the IG9 10yr itself, the one that allegedly CIO had distorted moved from +2 down to -1. But the “IG9 forward spread that no-one talked about while this was the target of Drew actually moved down in performance from +7 to -10 over the same period of time”. 17 basis points or regular underperformance for the “IG9 forward spread”, THE position of CIO…. This quite regular pattern on the”IG9 forward spread” in almost “no man’s land” markets week after week is really surprising all the more so as the IG9 10yr alone contributed only 3bps over the 17 bps drift in total….. It feels like the manipulation mostly hit the ‘7yr’ and the ‘5yr’ while the ‘10yr’ was almost unmoved. …And the IG9 7yr or IG9 5yr clearly were il_liquid… How can a market price that does not trade, is not even quoted as such, and is not liquid like the one of the “IG9 forward spread” move with such regularity? That required quite a fine tuned synchronization between indices that barely traded and quoted in very large “bid-offer” for small volumes anyway….How could it last like this for 3 months or more? It was sneaky for sure…

That mandated a bit of time of observation over several sessions to ascertain the conclusion. After 3 weeks though, only one explanation was possible: the market was rigged, targeting CIO “forward spread” exposures already and this is what Iksil already suggested to David Goldenberg no later than January 11th 2012. Goldenberg would not disagree with Iksil’s suggestion finding it an “interesting” point of view…May be he felt confident and “interested” because he had the IB of JpMorgan on his side already….This is what the evidence of this written chat would strongly suggest if at last it was publicly disclosed by the authorities…

As to the alleged “pressure of CIO” on price that really could only have happened in the last week of February 2012, the table here provides the observed changes in prices of the time in the circle. After trading a total of $13 billion over a week (of which $7 bln traded with Weinstein in 10 minutes just 40 minutes before the closing of New York at Weinstein’s initiative only) on IG9 10yr in a market neutral fashion mostly. During this week CIO did about 25 to 35 trades (ie 7 per day on average). Based upon ‘bid-offer’ being 2-3 bps wide in general, the resulting effect was a 1 to 1.5 bps outperformance of the IG9 10yr versus its peers and some 0 to 1 bp of outperformance of the “IG9 forward spread”. This “pressure” is measured here over a full week, not a day. This “pressure” resulted after about 30 trades, not just a couple of trades done in one day or an hour as Weinstein did

92 actually towards the new york close at February month end specifically so. More, as Iksil traded in a written form on Bloomberg chats, the evidence does exist to prove that AS the New York close on February 29th neared Iksil tried to trade a “better levels” for CIO ie levels for the IG9 10yr that induced a BIGGER LOSS for CIO year to date.

Thus the proof exists that Iksil deliberately and pro-actively tried to remove any potential pressure resulting from the former trades of CIO that he had executed. Here Iksil actually tried to trade for size in a way that blatantly induced A BIGGER LOSS for CIO in the estimate P&L. Thus Iksil’s behavior contrasted totally with the one of Weinstein as Roberts described it on a recorded phone call on March 1st 2012. Thus, not only there was no such “pressure” as it has been characterized in some media reports but also in the signed settlement of the Jp Morgan with the CFTC and the FCA (October 2013), but the proof does exist that Iksil INSTEAD behaved perfectly well while Weinstein did not in that regard. As said, every single regulator, the bank or every investigation team had knowledge of this proof here. They would ALL conceal the evidence and the facts from the public eye. They therefore conveyed later on a 100% misleading picture of the events that occurred then with regards to Iksil’s actions of the time. Yes they had had the written Bloomberg chats and call transcripts of February 29th or March 1st 2012 under their eyes proving how wrong they were in their public disclosures however ambiguous they were in late 2013.

Where was Jp Morgan standing as a whole market player here at the time in February 2012? Well the evidence does exist again that would show what the net exposure of the bank was. Why should it be left under confidential seal today in 2017? As explained formerly in this document and on this website, the bank made net a fortune through the event of the “London whale”, lost a smaller fortune at CIO alone, and did NOT have to unwind CIO exposures in the markets for most of them. One can infer only here what the net exposure of the firm could have been. This is what the coming table displays. This is just a speculative attempt to picture a very likely scenario for the end of a strategic hedge managed “pursuant to the firm’s objectives”. This would show well enough though why Jp Morgan as a firm would likely opt to “ignore” the activism of Weinstein on late February 2012 and support all along the plain mischaracterizations about “price pressures”: it was making a hell of a gain then and people would not suspect it of market manipulation.

The table below shows what the situation likely was albeit with figures that can only be seen as “order of magnitudes”. If one remembers, at the very start of 2012, CIO had a $51 billion long risk on IG9 10yr while the total “world active market” on this index was no greater than $35 billion for CIO….It was inferred from that, that the bank through the IB had at least an offsetting position versus CIO of $40 billion….The same reasoning applies to HY17 indices given the orders of Drew, as per the general balance of risks that Drew had commanded. The comment of Pinto to Macris about the IB “hedging CIO on High Yield and losing money in 2011” (call of March 23rd 2012) should be remembered….

The table here assumes that the IB likely has a net $40bln IG9 10yr short and a net $2 billion HY17 long risk on January 3rd 2012. ( all numbers in billions) Period dates IG9 outstand IG17- at IG17 HY17- ICE Bank Net Bank start: Jan 1st 10yr amount CIO Outstand. at CIO HY17 position Net traded amount oustand IG9 HY17 by CIO 3/1/2012 0 34.9 0 32.3 0 5.4 10 2 17/1/2012 9.9 41.2 -1.8 33.2 -1.7 5.6 19.9 0.3 31/01/2012 16 50 1 32.6 -2 5.1 26 0 17/2/2012 18 55 20.2 33.3 -4.8 5 28 -2.8?! 29/2/2012 31 61 23.6 32.4 -6.6 5.3 41 -4.6?!

16/3/2012 31.5 70 25.7 30 -7.4 8.5 41.5 -5.4?! 30/3/2012 34 71 33.4 33 -7.9 8.5 44 -5.9?!

This table has quite strong built-in assumptions whereby the IB would have kept more or less a constant exposure on HY and IG9 10yr all along the first quarter of 2012. This would lead to a situation where the bank as a whole owned at the end of the day $44 bln of net long risk exposure for a total market size of $71 billion. Same mad conclusion on HY indices by the way….Thus here Jp morgan would have weighed 62% of the whole IG9 10yr market. How credible is this? How likely the regulators or ICE could have ignored that monopolistic situation? Most likely here the IB was instructed to offset CIO’s trades day after day in the markets all along to remain at a net say $20 billion exposure for the bank ( ie CIO +IB+ “credit hybrids”). The IB would have done it maybe for “regulators’ optics” once again….As this was a “secret” the IB certainly would avoid meeting CIO directly in the markets. Hence the “middle man” would actually see this drama “JpMorgan-IB vs Jp Morgan- CIO” developing even further funnily enough (see the movie “Cramer vs cramer” for distraction- see also Macris to Pinto on the March 23rd 2012 call too actually). Thus in a sensible scenario, while the IG9 10yr position of CIO grew from $51 billion to $85 billion, the other offsetting positions in the bank grew likely from $40 billion to $65 billion or more via the IB traders. The move was not “so discrete” and the IB certainly “volunteered” an explanation for any market player asking “what the hell is going at Jp Morgan guys?”….

There was no candor in all this though as a simple phone call to Drew inside the bank should have sufficed in theory to stop this “marionette show”. No one did it right? Not even Dimon or Pinto or else among regulators….The March 23rd call between Macris and Pinto sounds like an attempt to calm things down when it is too late. And none of these top executives would need to ask anything to Iksil on the matter…They all know perfectly why this “marionette show” is happening then where Jp Morgan is “making a mockery” in the markets….It was way too late as Macris pointed out. The bank top executives had let the rumor grow since January 2012. And that had NOT been a “mistake” of theirs until Drew would make her gesturing “elevation all the way up”. That was not a “mistake” but something else that sounded like a clear motive to let the rumor against CIO grow and grow over the first quarter of 2012. As of March 23rd 2012, Macris actually told Pinto that this bad rumo had gone too far and that was why Drew was making all this gesture quite internally at Jp Morgan.

She apparently would not reach out to the regulators then, would she? No, and just no regulator would publicly blame her for that….Had she reached out to them she would certainly have claimed it for her defense later one. The case “Macris vs FCA” that settled in February 2017 strongly suggests that Drew had not reached out to regulators officially as she had NOT instructed Macris to “update the FCA” and he would not do that on his own decision. This remark unveils one very unfair aspect of this scandal…Indeed, one salient blaming question of the investigation teams shall be for Iksil to answer “why” he had never thought of contacting the regulators in late March 2012, precisely so. Well Iksil, despite his alerts and his frustration at times, he trusted his colleagues and the relevance of the orders he had received in the context of the “share buybacks of Jamie”. More Iksil through his internal training was told to reach out to his management and eventually to compliance with management prior approval. Regulators were out of his universe in that case. Now Drew was the one at CIO in regular contact with regulators by mandate. That was part of her duties in her position. She was elevating “very, very, very, very, serious accusations” to paraphrase Pinto on market manipulation themes. She was the most “accountable” person knowingly so at CIO. And she made all this noise within Jp Morgan. And, IF the regulators really are to be trusted on their crass “unawareness”, it MUST be that Drew had not warned them on March 23rd 2012. And that was a huge failure of hers. Now IF she has warned the regulators, this whole obsessive accusation against Iksil or the “CIO traders” is just a mere unfair persecution at best. And IF Drew

94 warned them, then she lied under oath before the US Congress in March 2013…She indeed misleadingly claimed having been “’betrayed” and have been “unaware of the price differences”….but she faced no charge anyway….

What about ICE now and other positions in the “tranche book” of CIO? Likewise the HY 17 figures provide the same conclusion. Thus while the market players saw CIO selling protection on IG9 10yr and buying protection in HY17, they also saw the IB doing the opposite trades all along the first quarter of 2012. Drew made all the drumming she could within Jp Morgan no later than the 23rd March 2012. Whether the blind and grossly “unaware” regulators ignored it, ICE saw it as well… And, as per the “London whale” official tale ICE would NEVER have warned either the Bank or CIO or any regulators, right? ICE was precisely here to prevent situations of this kind. So ICE warned even if every investigation team silenced it later on. ICE sent many warning reports to regulators, to Dimon, to Hogan, to Zubrow and else about this “JpMorgan IB Vs JPMorgan CIO” divorce. It is really impossible here to believe that such was not the case as otherwise the market share of Jp Morgan would have grown to mad levels on both IG9 and HY17 un-notified. And top executives made decisions about it no later than late January 2012. It is even more likely the case as in early February Braunstein (firm-wide Chief Financial Officer) and Hogan (firm-wide Chief Risk Officer) were directly involved in granting temporary but unlimited limit increases for CIO to keep trading like this on IG9 and HY indices via Iksil, now the “French trader wearing black jeans”…. They were fueling the bad rumor against CIO in plain knowledge, witness the ICE warning reports that came weekly at least since mid January 2012 most likely. It is worth reading again here the March 23rd call between Stephan and Artajo where the name of Ari Wechsman comes up….The IB risk guys did have a longstanding focus on the “HY vs IG” decompression trade in relation to the “tranche book” of CIO….This was no coincidence as well. More Barry Zubrow on February 13th 2012 had delivered a 68 page long letter to all the regulators and the US Senate about this “CIO tranche hedging book” in relation to the heated Volcker Rule. Artajo was secretly demoted at this time by Drew, Macris and Human Ressources department at Jp Morgan. Since not a single regulator later on would claim having investigated the matter then, it is very likely that they received as many “reassurances” as they wanted with regards to Jp Morgan not reaching too high a market share in IG9 indices or others….The “reassurance” must have been on the line of “the tranche book of CIO is being dismantled as agreed in the past”, this coming with a piece of “trader flesh” on the silver plateau….

One last table will summarize what happened and what Iksil kept describing throughout all the CIO ranks during the first quarter of 2012. As this table will show CIO did record an ongoing and regularly growing loss. This must be matched on the side of the IB and SABA among others by ongoing and regularly growing gains, and nothing else than that…. One will see in bold aggregate figures for every month of the first quarter and a total for the quarter itself. The IG9 “block” included some tranche trades on top of several different index trades that were meant to synthetically produce the “forward spread”. The figures here again are recollections that are meant to provide an “intuition” expressed in basis point terms on the IG9 10yr index price itself. ‘others’ is a label to include in one group of market players the IB and other hedge funds like SABA that traded in opposite direction to CIO then quite deliberately so. One can see that by a remarkable coincidence the “drift” of the IG9 forward spread mimicked so well the growing gains of “others” and the corresponding growing loss of CIO all along the first quarter. One would also notice the order of magnitude of the losses here generated by the sole “IG9 block”. This is a pattern that would be elevated inside CIO by Iksil from the first week of January 2012 onwards and through the daily estimate P&L reports done by Grout, just completing here the many repeated alerts and loss projections elevated by Iksil in all possible forums. The amounts at stake here were $700 to 800 million depending on the timing that “others” would have had on the period. They were making a hell of a gain against CIO knowingly so. This fully contradicts again the legend that Zuckerman and Rhule would start conveying misleadingly so in early April 2012 with their seminal articles. Zuckerman may been ignorant of that, but no Rhule for sure….

Some other orders of magnitude have to be highlighted here. CIO chiefs hoped to lose only $30 million on the “tranche book” at the start of 2012. The loss was already beyond that figure by the 10th January 2012… And Drew, MAcris, Artajo would then order to trade more and more and more. At the end of January 2012, Iksil alerted of a $300 million drawdown most likely to come by the end of the first quarter 2012. It was the 31st January 2012. Macris found it “worrisome” and demoted Artajo on the follow with the support of Drew and of Human Ressources. But Iksil would NOT be informed then while being order to keep trading and trading. Artajo of course but also Drew and Macris would issue these orders face to face to Iksil then. And the loss would keep growing at CIO. And the growing projected loss would be well elevated “all the way up” in March 2012…. And Drew, Macris, Artajo would then order to trade more and more and more

By the end of February 2012, the year to date loss was then beyond $200 million and Iksil then projected a $300 to 800 million potential further loss to surge by the end of March 2012. The projection was largely documented through slides that Iksil had designed and widely distributed to management, risk control and other colleagues for advice. This led to the emergency meeting call by Macris for Bacon’s plain involvement no later than March 2nd 2012. There would be this repeated “NY led” March 5th - 6th order to alter the estimate P&L process. There would be the March 12th 2012 meeting with Ashley Bacon and Ian Green to speed up the “offshoring” and internal collapse with “credit hybrids” of “tranche positions” in the firm. And the losses would balloon soon after as projected by Iksil for so long. Now this projected loss or “drawdown” in Drew’s words is “only” $800 million in total at this time, not $6 billion or more. It remains that the book was in “post mortem” no later than the 26th March 2012 and that the elevation of Drew was considered as a “very, very, very, very serious accusation” by Pinto (JPM UK CEO as of March 23rd 2012- 2 weeks before the seminal articles). Thus while the other $5.5 billion of losses at the CIO had not shown up yet, there was not an inch of room for complacency in a bank like Jp Morgan that already traded actively through the IB to offset the most recent “active” trades of CIO both on IG9 and on HY indices….Not a single $1 of the loss would be missed or “unexpected”.

Dates CIO IG9 “block” YTD ‘Others’ ‘Others’ IG9 fwd Outstanding amount trades estimated estimated trades P&L cum on IG9 price P&L impact “drift” 10yr ‘impact’ impact ($million) 6/01/2012 4bln 0* 0** ?? 0 $34 bln 11/01/2012 6Bln -1Bp +$25 ?? -$25 $37 bln 18/01/2012 3Bp -$50 -3bln +$50 $41bln 25/01/2012 3bp -$100 -2bln +$100 $43Bln 31/01/2012 6bln -1bp -$75 +$75 $50 bln January 16Bln 4bp -$140 -5bln +$140 4bp $50 bln 8/2/2012 5bp -2.3 bln 17/2/2012 1bp -$305 -2.5 bln +$350 10 bp $55 bln 29/2/2012 9Bln -3 Bp -$195 +$290 8bp $55bln (before close) SABA -1bp -$160 -5Bln +$255 7bp $60bln trade NY close February 15bln 2bp -$195 -10bln $290 6bp $60bln 7-8 March 4 bp -$320 -5bln +$395 11bp $70bln 2012 20/03/2012 1 bps -$360 0!! +$435 12bps $70bln 30/3/2012 9 bps -$720 0!! +$795 21 bps $70bln

March 3bln 13bp -$720 -5bln $795 19bp $70bln Q1 34bln 19bp -$720 -20bln $795 19bp $70bln  *The IG9 outperforms with the HY index and the IG9 5yr squeezes even more despite CIO buying $8.5 billion protection. The IG9 outperforms normal IG indices by some 5bps: 3bps come from HY rally and 2 bps are specific to IG9.  ** the HY performance balances the gains on IG9 positions

So it seems obvious that the bank, through the IB traded in order to keep its net market share on at least the IG910 yr and some HY indices under a certain level. The IB was therefore making larger and larger gains while CIO’s performance plunged deeper in the red. CIO was perceived as “protecting the depositors first”. Therefore CIO had quite a compelling “excuse” to keep trading. Of course the CIO also had a motivation internally embodied by Drew’s stubbornness in picturing her “marionette” on the front-stage in the markets. She would label it in January 2012 like Bacon would later in May 2012 on the lines of “regulators optics”. At the end of the day, this would be a standalone loss of $6.3 billion for CIO and some $50-60 billion net ultimate gains for the bank that would show quietly through 2012, 2013 and 2014. And no regulator would blame the bank for that as such. Had CIO gone into a mad trading strategy or had CIO simply followed orders coming from above Ina Drew’s head actually? How did those other $6.3 billion of losses at CIO, due to price changes in the markets, arrive? This is what the next part is going to describe. One may speculate that Drew was not “expert” enough given the complexity. One may also think of “hubris”, ego-mania as the part of u human beings that can be forgiven at times….There was some of it but it did NOT drive the CIO managers’ decisions.

Once again this is going to shave a lot of the candor that one could lend to the senior managers of the bank. Drew maybe was not the most stable person in the bank at the time (see Pinto on the March 23rd 2012 call again with Macris). But Drew was supported 100% by no less than “Jamie”, “Doug”, “John” (Hogan), Ashley Bacon and a couple of other “partners” all sitting at the bank’s operating committee (see email chain between Drew and Dimon of April 5th 2012 once again).

b. the loss analysis shows that the bank maximized the loss at CIO on skew related positions while trying to reduce the total loss for less clear reasons

1- What was the actual move? Answer: the forward spreads underperformed and drifted against all logic against all existing liquidity patterns

The focus will be set here on the first quarter of 2012 and extended until mid May 2012. Beyond that date Iksil would be kept away from the offices by HR staff until July 2012 where Iksil would be terminated by HR staff as well. The figures may not be quite accurate as they just reflect Iksil’s recollection as it was soon after the events. It remains that Iksil did that kind of analysis everyday in 2012 in the office and elevated the situation every week at least in a state of permanent emergency when the phenomenon occurred. Although not 100% accurate in the figures, the report here will provide the reader with a good sense of the directions and the main mechanisms that were at play anyway. The charts that follow show in a summary way what Iksil observed in the markets and elevated all the way up as of the 31st of January 2012 to management, finance, Human resources, risk control and risk modeling.

The chart below displays a situation that looked “normal” on the surface. One will notice that the “on the run” IG17, the now famous IG9 10yr and the most liquid HY index, ie the HY17 then, had been trending downward in spread terms. This was typical of what is called a “market rally’. This was the “rally” that “Jamie” had

97 anticipated in December 2011 and ordered to prepare the “tranche book” for. And thus, as the risks were balanced since 2010 actually, the estimate P&L reports should have shown relatively benign changes. As one knows today this was far to be the case. Indeed, while this market seemed to move as usual on the face of it, the CIO “tranche” book, which was balanced, recorded growing losses from the very first days of the year 2012.

The loss year to date grew at $30 million by the 10th of January. That could NOT be explained by the remaining $20 billion net short risk that had been inherited from year end 2011. And although quite high, the unwind costs could not explain this loss either. And the loss would keep snowballing “off the charts” then at $130 million as of January 31st 2012, then at $220 million as of February 29th 2012, then at $585 million for the 30th March 2012, $2.1 billion as of April 30th 2012 and close to $4 Billion by the 15th May 2012. As some articles pointed out in early June 2012, the loss was about $6 billion already then.

Only then, mid June 2012 the IG9 10yr skew was “settling” at zero for a while and the “tranche book” at last would be transferred… first to the IB books…That was so simple to stop CIO since 2011 “in hindsight”! .And here one really struggles to understand why this preliminary and so simple transfer to the IB had been delayed since the 26th March 2012, namely THE day when the “tranche book” was officially in “post mortem”.... And one struggles to understand why this simple internal transfer from CIO to the IB had not been done since late 2010, ie October 2010…. This “post mortem” officially surrendered the “tranche book” to the firm by no less than Macris and Drew as of March 26th 2012. The context pushed for this easy transfer to be done “at once” beyond the sole action here of CIO top chiefs. Ina Drew had “freaked really”, and would allege later that she had ignored the existence of those “trades” and “price differences”. She had ordered to “stop trading” allegedly so too. More the march 23rd call of Pinto and Macris secures the fact that she was then uttering “very, very, very, very serious accusations” and that Pinto knew of the price differences and that Hogan knew of the trades that Drew apparently ignored, the very same trades by the way that had blown up just all the risk limits of CIO. The massive breaches of CIO limits were elevated to Drew and the whole operating committee if one wonders… Really, who other than “Jamie” could have postponed further this seamless and planned preliminary transfer to the IB from March 26th 2012 till mid June 2012? And why while the reputation of Jp Morgan was clearly at stake ( see again the March 23rd statement of Macris to Pinto “JpMorgan is making a mockery”)? It must be that something even worse hung as a Damocles sword….

There might have been the scenario where the “fighter” Drew wanted to make this transfer difficult in the backstage… How credible is this as of March 23rd 2012? Lets’ review the facts here. Since February 9th 2012 Drew had opened the talks with the IB in person (Grout would tell Iksil here who did not know before). As of the 23rd March 2012 she saw her CIO in breach of all the main risk limits and this time this was truly due to the changes that she had ordered for the “tranche book” since early January 2012 to “maximize P&L” and manage “regulators optics”. What is the result? Against a budgeted loss for the whole 2012 of $30 million, the “tranche book” had already lost $250 million or more. She knew since December 2011 (see the April 17th 2012 email above from the OCC summarizing what she had told the OCC herself) that she could NOT unwind the “tranche book” positions in the markets. She knew that regulators expected this same “tranche book” to be taken down. She also knew that “Jamie” projected about $60 billion of quite tangible capital gains from this so simple transfer from “CIO to the IB” of the very same “tranche book”. She thus had every reason to surrender and ease the transfer to come towards the IB. She by the way made a hint to that in the April 5th 2012 email chain displaying her surprise to see “Jamie” being on vacation apparently. She then had been waiting for the “go ahead” of HER boss to proceed with the transfer. HER boss was Dimon and no one else.

What she fought for was maybe the “price difference” actually that she would later allege ignoring. She was after $300 million on the face of it. The IB collateral team was then in the second week reporting to

“its supervisors” that every day it had adjusted CIO “estimate P&L” London based prices for an amount that had kept growing and was then beyond $500 million in total (see Mark Demo emails around April 20th 2012 on the matter among the US Senate report exhibits). So if really Drew had been “unaware” of this price difference and fought the timing of the transfer based on that P&L difference (see the email of Artajo to Drew and Macris on the matter dated March 23rd 2012 too) accusing the IB “very, very, very, very” seriously, for one the IB collateral staff would let her know of the $500 million difference, and for two one wonders on which other grounds than price differences she would have accused the IB so seriously. She had no reason to fight that transfer herself. So one is left with “Jamie” as the one in command of this very surprising delay for this so simple and “cathartic” “internal transfer”…. Now, as per the Sarbanes-Oxley laws, such internal transfer had to solve the “price differences” elevated by Drew for good…

After the commotion made by Drew, the transfer could be processed even if the “price difference” dispute inside JpMorgan had not been solved yet between CIO and the IB as such. There would have been a $300 million loss pending. Actually CIO could have shouldered this easily… Provided the issue was just that actually… What could have been therefore the rational reason of “whoever at JpMorgan was in charge of this CIO-to-IB transfer” since there must have been one reason at least?... with a “mistake” again or with no “mistake”… It remains a mystery in 2017 for sure. But clues to the answer do exist. The short answer again is : massive liquidity reserves had been missing since 2007… regulators and top executives feared the Pandora box standing behind these $300 million of price differences…the positions were very il-liquid since 2010… Thus it is worth looking at this chart below first to measure that nothing major or material occurred in the markets through the London Whale event in overall price changes……To be clear, the “London whale” event was a massive market manipulation but around global levels that would not move much and thus, it had nothing to do with a crisis scenario, or extreme price variations albeit short lived or a “unique distortion”…..

Just to give an idea of the order of magnitudes involved through the “London whale” event: the IG indices moved by say 50-60 Bps and the HY indices moved by about 200-250 Bps. They moved grossly speaking according to their long term anticipated volatility on a standalone basis and relative one to each other. The HY index indeed is known to move in a ratio of 1-to-5 when the spread ratio itself is of 1-to-5. This is why in the chart above, since the HY scale on the right is adjusted to the IG scale one can see lines moving broadly in parallel. The scope of the spread change was also relatively benign, actually quite similar to 2011 and 2010. If one wants to compare with 2008 and 2009, which were years where markets were really moved by events indeed, one can remember that IG spreads moved between 60 Bps and 350 Bps (thus almost 300bp for the range instead of 50 Bps) and HY spreads moved from 600 bps to 2500 bps (ie a range of almost 2000 bps vs 200 Bps).

All this does NOT explain why the CIO “tranche” book kept losing money so much in the beginning of 2012 then. None of this would justify the delay between March 26th 2012 and mid June 2012 of this “CIO to IB transfer” of the “Tranche book”. The next chart will display a better picture considering that indeed all those spreads were expected to move identically in proportion as history and common sense showed day after day since 2004, the year when the credit indices took off as traded products.

As one will see, the historical pattern was ALMOST respected but not quite. The change from 2011 was abrupt although barely noticeable except for the CIO at the time. Not a single public report will make reference to that. The shift is circled on the chart below and the only plausible root cause is provided: after the March 12th 2012 meeting with Ashley Bacon (deputy firm-wide Chief Risk Officer), CIO had already surrendered the “tranche” book to Dimon for the “exotics credit derivative wind down” to proceed at last. It was not even CIO chiefs who resisted the “transfer”. In practice, Pinto and Macris would call it an “externalization”. “That will happen” certified Pinto on the March 23rd 2012 call in front of Macris who seemed to have doubts… Pinto, the JPMorgan CEO of the UK subsidiary of the US bank, was assertive and reassured Macris on the matter.

And one really puzzles again about the “cause” for the delay. All the more so as this “transfer of CIO to IB” was quite simple, agreed and planned to happen like “very soon” according to Pinto. The FCA would call it an “off- shoring”. Blue Mountain, the hedge fund involved, would call it a “facilitation” as far as it was concerned. What was it in practice? It was to transfer the “tranche book” of CIO in the end out of the usual perimeter of the bank balance sheet to a hedge fund like Blue Mountain so that the positions of CIO would be collapsed only then, through the hedge fund, at a certain price with “credit hybrids” legacy positions. The operation would “de- consolidate” the CIO exposures as decided by Bacon as of March 14th 2012 officially inside Jp Morgan. Then a “third party” would control the valuation as per the plan of Dimon and as per the requirements of the regulators for them to “approve” such transfer.

Here on the coming chart one can see that suddenly the IG9 10 yr stopped radically to follow the path of the IG17 and actually massively underperformed the HY17. It matters to specify that CIO had stopped trading actively the IG9 10yr since the New York close of February 29th 2012. That was thus 2 weeks BEFORE Bacon’s move here. Thus this brutal shift is not due to any pressure imaginary or real attributed wrongly to CIO trading activity anyway. This is completely abnormal and a clear new indication of the manipulation that would occur next. Weinstein, Hubbard or even Vilani would have nothing to do with that. It thus did NOT start with the articles of April 6th 2012 but with the March 12th Ashley Bacon meeting as far as the IG9 10yr is concerned versus its peers and against some 7 year long historical record. And any alleged “pressure” of CIO had vanished already…

Yet, as the next chart will show, this brutal shift), was not at all a surprise for top CIO managers. As they could well anticipate it was coming from within Jp Morgan meetings involving Chief risk officers at the firm-wide

100 level on behalf of his “exotic credit derivative wind down“ plan. This “wind down” plan was quite different from a “market unwind” (- see the April 5th 2012 email chain between Drew, Dimon and the operating committee). The next chart will thus show the same kind of figures, ie proportional moves but for a bit more indices. Here will be added the IG9 5yr index that was one component of the “IG9 forward spread” that also included the IG9 10 yr index. This “IG9 forward spread” trade was actually the one trading strategy studied by Drew between March 2011 and June 2011, approved by her in July 2011 and ordered to grow next by the same CIO top chief. The ratios and components had been shaped by Evan Kalimtgis, Keith Stephan and Artajo back in 2011 (not Iksil) so that CIO risk metrics would be improved especially in stress scenarios analysis. The chart below starts from the beginning of December 2011 actually because this is how Iksil could spot what would be known inside CIO as “the drift” on the IG9 forward.

The IG9 10yr index path deviated from IG17 path right after the Ashey Bacon metting of March 12th 2012

One can notice that the HY index only reduced by 20% while the IG17 reduced by say 32%. This was relatively normal then as a risk of default was relatively getting much larger in HY indices when compared to IG indices in a lasting rally. One should notice ALSO that the IG 9 10yr followed quite closely the path of the IG17 until early March 2012 which was quite natural given that the IG9 10yr had mostly a similar IG profile. The IG9 10yr had some components that were considered “HY” but that was priced already in the spread difference versus the IG17. The divergence would not be caused by Weinstein and Al. It would be done only after March 12th 2012….And the CIO chief had no doubt this would happen once they had called Bacon in the look by March 2nd 2012: CIO would be hammered down in P&L….And the CIO chiefs woud convey instructions on March 5th 2012 and again on March 6th 2012 in anticipation of this coming hammering…And they would “elevate all the way up” the effect of this hammering as of the 23rd March 2012 ordering to “put the phones down” in a complete gesture that did not fool Pinto then…

100

101

As pointed out, the CIO top chiefs knew quite well what was coming since the very first days of 2012. The circles above spot the 3 key dates when Iksil elevated his concerns about the targeted manipulation of the markets against CIO exposures. One must admit that this manipulation is almost invisible in comparison to the actual magnitudes of the market changes. Those broad market changes were pretty normal as such. The brutal shift triggered by the March 12th 2012 meeting with Ashley Bacon was quite visible in the detailed analysis of CIO’s performance in its “tranche book”. But here on the charts above one struggles to detect it and needs a magnifying lens for the period preceding Match 12th 2012. Yet the move had already started by then. This is what the alerts and the slides of Iksil would describe all along.

What matters is to follow as best as possible the “IG9 fwd chge” line ( continuous light green) and the “IG17% change” dotted violet line. Based on month end November 2011, the IG9 forward positions had outperformed the IG17 market into year end 2011. CIO was NOT trading though if only to unwind very small exposures on IG9 then. By the end of January 2012, one may finally succeed in noticing that the “IG9 fwd” line traded almost back to “flat” versus the IG17 % change line. This means that the IG9 forward exposure had lost a couple of basis points, where one single basis point was worth about $30 million each….CIO did trade often but did NOT impair the underperformance of the its IG9 exposure then. This slight but certain underperformance could have easily explained why this “balanced tranche book” had lost still about $100 million in January 2012.

Actually Iksil explained and Drew, Macris, Goldman, Kalimtgis, Stephan all “got it”. What was the ‘factual’ explanation for that underperformance of the IG9 forward while, as said above, the IG9 10yr itself followed still almost its long term historical pattern? Well the blue continuous line shows the “rational” reason: the IG9 5yr had crazily outperformed the whole market but only from the 23rd December 2012. This is stunning as the IG9 5yr index was very il-liquid and contained default risks that were massively underestimated already by market players. Why were they exaggerating their error of judgment here? They would still persist… The case of MBIA later in November 2012 will prove quite clearly that their speculation was plain wrong… but only in November

101

102

2012. How could they ever have such lucidity about their own mistaken speculation? Were they just “lucky” or were they “re-assured” by Jp Morgan about the CIO imminent exit? For a very illiquid index, understating a default risk massively, to outperform the whole market in a rally like this, that smelled like at least a “targeting” of CIO exposure on “IG9 forward” but in the backyard. Here there could be just NO alternative explanation supported by “market activity” since there was almost NONE on the IG9 5yr anyway.

Still, the slight but quite regular under-performance of the “IG9 forward” driven by this mad outperformance of the IG9 5yr would go on through February 2012. The CIO chiefs were already aware and concerned by that due to Iksil repeated reports on the phenomenon throughout January 2012 already. The slight “drift” in 2012 from long term history revealed a strange phenomenon that Iksil explained in late January 2012 in full indeed on all possible forums. Iksil was fully understood then as of 31st January 2012. Artajo would be demoted as a result by Macris, Drew and Human Ressources to secure that Artajo was to only focus on the “tranche book” of CIO from now on…Rather than stop trading on the follow they instead forced Artajo to order Iksil to keep trading. They had a plan “B” in mind here for themselves that would lead to this gesture on March 23rd 2012 of theirs.

Iksil produced charts about this tiny, daily and almost invisible effect on a regular basis inside CIO that had slowly emerged BEFORE the March 12th 2012 meeting with Bacon. The “drift” turned out to be extraordinarily regular in all this noise, all this lack of liquidity and amid these gross abnormal moves as the chart below shows

One can see here again that, after the key March 12th 2012 meeting with Ashley Bacon, the “IG9 forward” will drift- as if it was a “dive” in fact, ie much faster, but simply because then it followed the path of the IG9 10yr index in its massive and absurd under-performance versus the IG17. Thus this “drift” elevated between January and February 2012 was just the precursory sign of the massive loss that the meeting with Bacon would catalyze. This massive loss would thus really start post the March 12th meeting with bacon that was meant to further the

102

103 plan of Dimon about his “credit exotics wind down” operation. One will notice that, as of March 12th 2012, the outperformance of the IG9 5yr could not go much further. Thus this meeting with Bacon was sort of “well timed” to make the losses at CIO go even bigger to another kind of magnitude through a new kind of manipulation. In that sort of consideration, the articles of April 2012 and the misleading statements of Dimon on May 10th 2012 will just be other catalysts to make the CIO loss go even bigger. The chart above thus displays the first 2 compounding effects: the IG9 5yr led “drift” of the “IG9 forward” versus the rest of the planet, and the “brutal shift” targeting the IG9 10yr alone after March 12th 2012 Bacon’s meeting. How regular it was since mid December 2011! It matters therefore to focus upon what Drew and Macris knew while they repeated to stubbornly their trading instructions in January and February 2012….

2- the loss breakdown: trading costs, risk balance between drift and decompression, Ashley Bacon meeting (March 12th 2012), Dimon statements (May 10th 2012)

Until early March 2012, the year to date loss was actually quite “benign” when one considered the embedded trading costs that Drew’s order had induced knowingly so. The IG9 drift against its peers was more than balanced actually by other existing positions. The real issue was the execution costs of 2012 as ti will be shown in what follows.

CIO chiefs were aware of the liquidity issues, being worsened by the high visibility of CIO exposures since 2007. Again one wonders what Drew wanted here since 2011 other than plant Iksil on the scenery of the “trading scandal” in the making. There is no other plausible explanation for her stubbornness of the time especially in early 2012. The CIO chiefs indeed knew in March 2011 that it was not possible to unwind the positions in the markets unless CIO incurred massive losses in execution costs. A project to collapse positions internally was launched already in early 2011 at the firm-wide level since this liquidity issue was very well known (see the ‘off the run’ rule reported in the US Senate report that dated back from March 2010 and concerned the IG9 indices already) . The “strategy 27” had been created on the “tranche book” at CIO in July 2011 in the anticipation that some small parts may might remain at CIO for them to expire over time, not be unwound anyway or be wound down by other means. In late December 2011, after repeated requests made by the Federal Reserve to estimate the unwind costs of this “tranche book” of CIO specifically so, Drew and others knew from Artajo in writing that unwinding only the first 25% of the book could cost $500 million (normal conditions, not forceful unwind) and may actually lead even to a larger loss for CIO going beyond $1 billion in total. By the 18th January 2012, Drew went in a fury as she learnt from Iksil that actually the basic cost was not $500 million any more but rather $700 million (Grout made the calculation at Iksil’s request here). Drew was angry as she had just ordered Artajo on January 10th 2012 ( see US senate report exhibits on “maximize P&L”) to tell Iksil to stop unwinding and for Artajo to work on how to “maximize P&L”. If the “unwind costs” were damn high, so were the “execution costs” anyway. That is just basic common sense that could not escape Drew’s judgment at the time.

More, as she knew quite well as very experienced trader herself, when the positions would grow, the future trading costs would for sure add losses in the estimate P&L over the days. How was it the case? Well, the markets on IG9 or Main S9 indices were il-liquid. This meant that Iksil would NOT trade exactly at the price that Grout would pick for the estimate P&L reports in the future post the trading day itself. Iksil’s traded price, although a key level for Grout day to day, might be used on the day the index had traded indeed, but not for long next. The day after, Grout would use also other prices and over time Grout would be close to some “mid prices” that were NOT the prices equivalent to where Iksil had traded in the past. Iksil could only trade on the bounds. And the bounds were large in 2012. And Grout’s future price selection would induce a loss. This loss came from the difference between the price where Iksil had traded simply because Iksil traded on bounds, not around “mids”, and the price where Grout would be most of the time i.e close to “mids”. This difference in prices was

103

104 in general quite small, like 1 Bps for “still liquid” IG indices. But it accumulated over time as the drift occurred and the traded volumes increased as per Drew’s orders, as per Macris’ orders, as per Artajo’s orders all along the first quarter of 2012.

It is worth making a numerical example to secure the understanding of the phenomenon that every experienced trader knows. Anyone has actually come across that phenomenon when reselling a car, reselling a house, or reselling any good in a secondary market in fact. Unless one is “lucky”, one will not sell at the price that was originally announced as the “mid” price where it “should” trade. Let’s say for example that the IG9 10 yr quotes like “120 bid-123 offer” all along. Iksil will trade at 120, ie on the bid as prescribed by the approved “IG9 forward spread” strategy. Grout may use the price “120” on the traded day. But Grout will also have to synchronize this price with say the IG9 5ry and other tranche prices and other IG prices, and other HY index prices, and other HY tranche prices where IKSIL most likely would NOT have traded. Grout will thus get rather close to “121.5” which is the “mid” for this “120 bid—123 offer” quote. May be Grout was at “122” or he rather was at “121”. But he was NOT at “120” the day after the trade. This is at least what Grout did during January and February 2012. Thus over time Iksil will have traded by selling at “120” and Grout will value the trade at around “121.5” which induced soon a loss of 1.5 Bp in the estimate P&L albeit with a delay of one day or two. Indeed Iksil actually “sold” at 120 while soon after Grout would estimate that Iksil should have sold at “121.5” ie at a higher price instead. This “trading costs” or “execution cost” was reflected in large part over time in the estimate P&L that Grout produced day to day in January and February 2012. This phenomenon was monitored closely by Keith Stephan and his team through a report called “new trade P&L” that Grout produced along with the estimate P&L report.

The CIO knew since 2007 that the “CIO tranche book” lost every year about $200 million to $500 million in “trading costs” like the one that has been described above. This puts in perspective the net gains that were recorded all these years through the estimate P&L (between $150 million and $950 million between 2007 and 2011). Behind those net gains there was every year about 50% of those gains that had been received as “execution fees” by the counterparties of CIO. The “tranche book” of CIO was considered to be a “very good” client, ie very profitable for the market makers. Trading the “tranche book” in CIO had notoriously been damn expensive from the start since late 2006. Thus none of this “trading cost” issue was new at CIO starting the year 2012. Still Drew, Macris, Artajo forcefully pushed Iksil to trade in the markets while Iksil kept elevating the issues. One really wonders how Drew could claim under oath that she had been “betrayed” by the CIO London Front Office. One is puzzled to see that all the investigation teams involved would endorse this misrepresentation.

How much did those trading instructions of Drew, Macris and Artajo cost? Before looking at the estimated figures, one should refer to the US Senate report exhibits where it is seen that the firm-wide CFO Braunstein granted an unlimited by temporary extension on CIO RWA by February 3rd for the trades to proceed. It matters also to notice that the firm-wide CRO Hogan granted an unlimited but temporary extension of CS01 limit for CIO no later than February 9th 2012. Hogan also facilitated an artificial reduction of CIO RWA by supporting the IRC/CRM split no later than February 9th 2012 as well. Thus this trading flurry of costly activity was ordered by CIO top managers and was fully backed—on a temporary basis--- by CFO and CRO at firm-wide level. Once again this is when Zubrow (chief of compliance and regulatory affairs) sent this 68-page long letter on Volcker rule, citing this “CIO tranche book” that was about to die soon. It is really hard to believe that Dimon was not involved while HE was the one ordering to stop sending the P&L report of the IB to regulators right then ( see the US Senate report account).

104

105

February YTD loss $220 million

January YTD loss $130 million

The chart that above displays an estimate of the trading costs that were induced by those orders emanating from Drew since she had ordered Artajo to “maximize P&L” no later than the 10th January 2012. What a weird strategy to “maximize the P&L” of the “tranche book” of CIO if such was the aim pursued by Drew in this email! She maintained her stance here through Macris and Artajo with the direct support of Hogan and Braunstein in early February 2012. As a total those trading costs amounted to some estimated $260 million by the end of March 2012, about $80 million for January 2012, another $120 million more for February 2012, and $60 million for March alone. This is just an estimate but it is good enough. It is then interesting to compare the actual estimate P&L for January 2012: a loss of $130 million, including therefore an $80 million loss in trading costs. If one adds the $10-20 billion incurred from actual unwinds that Iksil did between the 4th and the 9th January 2012, one sees that $100 million of the $130 million total estimated loss came from either “unwind costs” or “trading costs”. If one recalls that the “tranche book” had lost $70 million on the default of Kodak around the 20th January 2012, one arrives at a total loss of $170 million versus a total $130 million reported for January 2012 month end. That has been no criticism for the price selection of Grout for January 2012. So one must assume that Grout was not that far from the “consensus” anyway then. One must conclude that the loss flagged on the IG9 forward spread was actually well balanced by gains on other positions held inside the “tranche book” itself. It was actually more than well balanced.

Likewise for February 2012, the month loss was $90 million (the YTD was $220 million) while the trading costs are estimated at $120 million. Thus, absent those trading costs incurred by Drew’s repeated orders with full temporary backing of firm-wide CFO and CRO, the “tranche book” of CIO in February 2012 was rather ok with an implied gain of $30 million overall net of trading costs. Once again the drift loss was quite well balanced by other gains inside the “tranche book” of CIO. Actually the positions had gained despite the “trading costs” and despite the further drift on the “forward spreads”. The maths are super simple here. Drew could do it in a split second. If not, Stephan or Kalimtgis could. They did actually as part of their routine job. Even Goldman could while allegedly “learning the ropes”. Still she persisted ordering Iksil to trade and trade and again… One may start guessing here the second thoughts of Drew as she knew the “Tranche book” was dead anyway. It was actually fairly balanced aside from “regulators optics”…..She may have meant to show internally that absent “unwind costs” or “trading costs” the book had a flat performance versus the markets. So would be her initial excuse back then to address the question issue about the missing reserves….

105

106

This induced that this long planned Dimon’s “credit exotics wind down” plan should NOT be finalized at CIO’s full expense but at the firm’s expense in part at least. The price differences were peanuts here. One knows that the real figure that they all had in mind was NOT say $500 million but rather $6 billion as elevated in late December 2011 through the Federal Reserve requests in fact. As such the alerts of Iksil bearing upon an imminent $300 to $800 million drawdown were actually “small fry” here. The alerts were accurate though as the events would prove but they were NOT relevant in the context. A lot more was going on. This still mandated a reserve to be taken against this predictable “drawdown”. This reserve too shall be missing right?

When one understands that CIO top chiefs and bank top chiefs knew in late February 2012 that the Year to date loss was in fact ONLY due to “trading costs”, one wonders why things got so bad in March 2012. There was no reason to postpone one long planned transfer that would lie in the sole hands of Dimon at Jp Morgan. Of course regulators had to “approve”. Why would they have blocked such a sensible move even if $300 million were at stake? From that standpoint, one wonders why the “London Whale” event would ever come to the light of day. Well the very sure thing is that they wanted Iksil to keep trading and trading again and again.

What was the change in March 2012 actually that likely “got out of control” in Dimon’s words? In short: Iksil stated on March 1st 2012 that he would stop trading even if that implied that he was fired by refusing to comply with the orders of Drew. On the follow Macris took an initiative then that was effective no later than March 2nd 2012. Macris had brought Ashley Bacon straight in the coming evolution of the “tranche book” of CIO. Some may believe here that Macris openly doubted Drew’s current management of this coming “transfer from CIO to the IB of the tranche book”. They may be right: Macris wrote deliberately about his doubts as to their ability at CIO to “defend the positions” given the RWA “commitment”. Macris and Kalimtgis discovered a lot of hidden things themselves right after the “CIO Business Review” of 29th February 2012…Kalimtgi would resign right after that CIO Business Review “discovery” of his. This radical move of Macris arrived at the very same time and has been vastly under-reported in every single public report.

This move of Macris then may look as a “good thing” on the face of it. But once the action of Iksil is known in the context of those huge trading costs, one sees differently the initiative of Macris to call Bacon in the loop as of March 2nd 2012. CIO had surrendered in the most visible way it could…Yet the decision to start the “CIO to IB” transfer had always been in the hands of Dimon at JpMorgan. Regulators would have their say too….

This initiative of Macris will lead to first this repeated “March 5th -March 6th order” from Drew to amend the estimate P&L process but WITHOUT updating the NBIA of 2006. Was Drew mad or was she “covered” here? This initiative of Macris shall ALSO lead to the critical March 12th meeting with Bacon. Why did Macris make such a move while actually they at CIO had the backing to CFO and CRO firm-wide? Well, again, Iksil had stated that he would refuse to trade further as of March 1st. That likely was THE problem and neither Drew nor Macris wanted to become the next fuse: il-liquidity and visible absence of “active trading” were too loudly calling for liquidity reserves that had been missing for years already. Iksil could not be fired because he had brought up compelling evidence. He had come up with reports to Artajo based upon the call of Gabriel Roberts (CITIGROUP) among other things that CIO was targeted, that there simply was NO market on IG9 10yr. This alone is a blatant rebuttal of the bank statement, backed by just all the authorities in 2017 though, that the IG9 10yr was “not il-liquid”. Well it was il-liquid and manipulated. And Iksil had widely distributed his latest slides predicting a loss for March 2012 alone ranging between $300 and $800 million. The reserve for drawdown could not be avoided any longer. And that would be a genuine Pandora Box here for anyone who raised the matter at the top of the bank towards any regulator.

106

107

Iksil on the way characterized the IG9 10yr position as “scary”, huge and the like. It was carved in stone by Iksil who refused to keep trading at the risk of being fired….Macris had to move fast and radically indeed for the “transfer” to have a chance to occur by the end of March 2012 as far as he Macris was concerned. But things could NOT go that fast despite the official “post mortem” stated as of March 26th 201. That was the core problem that would spark the “London whale” manipulations in the media, in the markets, in the accounts….

3- The peculiar synchronization of losses across the book after the Bacon March 12th 2012 meeting

It is time now to look more accurately on the effect of this March 12th meeting. Its effects were precursory of the “trading scandal” to come. This distortion on the IG9 10yr would not come alone and certainly would alert the market players even more. It really was gross and visible for any player involved day to day in those markets of credit indices. Jp Morgan was weighing on prices against its CIO…So was the rumor in 2012… At CIO, the manipulation was even clearer as it will show in the coming charts. Indeed, from then on, the losses will show just everywhere in the “tranche book”. So far, as one could see, there was a growing loss on the IG9 forward spreads. But, up until March 12th 2012, once one took the trading costs in consideration, really the year to date loss came from trading costs induced by Drew’s orders alone in fact. The book was balanced overall outside of this cost. The IG9 forward regular and nonsensical underperformance was balanced by other positions, especially another strategy called “decompression”. The book indeed had had this structural position, as wanted by Dimon since May 2007, whereby the book purchased protection on HY markets and found a way to finance that protection using IG markets. As seen before the HY performed by 20% while the IG market performed by 32%. This brought a gain day to day that balanced mostly the “drift”, no matter how weird the drift was by itself. This gain was consistent with history and current market bullish moves. But as Iksil refused to trade, as Macris had deployed his defying initiative with Bacon, this march 12th 2012 meeting would start making all the positions other than the “IG9 forward” one to lose money in an extraordinarily well timed manner.

Let’s see first the IG9 forward spread with a chart starting as of the end of November 2011. Why this date? Because this is the moment when “credit hybrids” was closing its tranche business. This is when Bacon was in charge of moving the legacy tranche position of the IB to a hedge fund like Blue Mountain. This is also when the IB and CIO had different prices that caused a valuation difference for the CIO tranche book of $300 million or so. Thus this was likely the reference point for Bacon and the firm. The chart below will show that as of March 12th 2012, the drift so far in 2012 was just compensating an initial outperformance of the “IG 9 forward” in late 2011 versus the market while CIO was NOT trading other than trying to unwind actually. As the coming chart shows, the effect of Ashley Bacon involvement leaves no doubt. One can see that the April 2012 press articles were peanuts in comparison as triggers. They simply were the further catalysts of something that had started on March 12th 2012 with Bacon’s direct involvement.

107

108

The following charts will show what the risk controllers reported to Drew soon after on top of Iksil’s alerts and independently of Iksil anyway. The next chart displays the evolution of the main driver of the estimate P&L report, ie the overall loss caused by the index positions. Indeed, there were positions on tranches and there were legacy “residual” positions. But they weighed little on the total figure. The chart below is thus not expressed here in Basis points but in $ million. The total index P&L cumulates the “decompression trade” that balanced the “IG9 forward” exposure and the “Itraxx Main S9” exposure (similar to the IG9 one). One will see for example that as of April 30th 2012 the total Loss on those principal index positions was of $1 400 million while the total reported estimate P&L for that date would be about $2 100 million. Thus 2 thirds of the loss came from those largest index positions. The remaining $700 million or so came from the tranche positions and the residuals. It will be seen a bit later how they compounded. But let’s first look at the index P&L where the IG9 10yr position was along with what was effectively balancing the sneaky “drift”…..

108

109

The drift is balanced March 12th 2012 meeting effect

To be sure, as one could guess, the long term historical relationship between HY and IG indices in the US broke down, right after CIO had met with Bacon on March 12th 2012 as the next chart shows.

109

110

Now one wonders what happened to the other exposures, being residual and very illiquid in any event. The next charts will show a perfect synchronization of losses after March 12th once again….. First the residuals (P&L on right hand scale) versus the main Index P&L total

One may notice the absurd outperformance of the deeply il-liquid indices like IG14-IG15-IG16 during January 2012 versus the more liquid indices right when the markets were rallying. This is another non-sense.

As to the tranche exposures, the next charts will display the timing of losses….Here the charts and figures are really proxies. Only Grout could have provided the proper path as the tranche P&L calculation over time was complex and involved data that only Monster truck (the CIO P&L secured valuation engine) could provide. Iksil could NOT access Monster Truck in practice as explained before. And Grout never provided Iksil with proper P&L records on tranche despite many requests of Iksil during the first 5 months of 2012….Yet Grout would provide the figures to his boss Artajo in February 2012 on Artajo’s request only in preparation for a meeting with Drew. All this would be done by Artajo and Grout without Iksil being in the know. Iksil will discover this by chance sometimes in 2016….As the next chart shows, the March 12th 2012 meeting will actually speed up the trend while CIO was almost NOT trading on tranches then. One can see that the articles had a counter- intuitive effect that will revert thanks to the ambiguous statements of the bank later in April 2012.

110

111

If one now wants to look deeper in the details of the shift, one can see about the tranche positions that referred to the IG9 index itself in the chart below what the effect of this bacon March 12th 2012 meeting was:

111

112

One may also be curious about what the effect was on tranches on the ITRAXX Main S9 index was while this was neither a strategic position for the firm, while this was an almost dead market, while this did not matter much in P&L terms for the bank:

On can observe the same effect and next a counter-intuitive reaction of the markets in April 2012. But Dimon’s misleading statements on May 10th 2012 will secure that this will be a loss too for CIO anyway.

What about the HY tranches that did not trade since late 2011…..

112

113

The “plunge” around mid March 2012 is blatant once again. Anyone who ever tried to test an inactive market and saw whether the quotes would bulge? Well an inactive market just does NOT care about whatever you would say, do or try. This HY tranche market was NOT trading any longer since December 2011…. But it moved after the secretive March 12th 2012 meeting that occurred inside JpMorgan’s fortress walls….

Thus the March 12th 2012 meeting made the loss balloon across every single exposure of the “tranche book” of CIO be that residual index positions, “IG9 forward”, ‘ITRAXX Main S9 forward”, “IG9 tranches”, “ITRAXX Main S9 tranches”, “HY tranches” etc….. Therefore it was shown here that the position analysis had gone to the finest possible details inside the firm way, way beyond a level of scrutiny that people usually look at when monitoring their strategy. The surgically spread loss after March 12th 2012 is proof of that. The effect of the “drift” and the March 12th 2012 disruption on IG9 10yr had been flagged already in due time. Next it was shown that the main drivers behind CIOs moves were the “trading costs” inherited from Drew’s instructions.

Iksil’s refusal to trade as of March 1st 2012 plus his alerts, and the “initiative” of Macris to bring Bacon straight into the final stage of the long planned “exotic credit wind down” plan of Dimon would be key. The “transfer” should have occurred straightaway. It did not happen before mid-June 2012 actually. In the meantime, this initiative of Macris would lead to the highly controversial March 5th-March 6th order and the systematic-surgical hammering of CIO’s positions next from the 12th of March onwards. Thus Dimon completely misrepresented the situation on May 10th 2012 alleging that those positions of CIO had been “poorly executed, poorly reviewed and poorly monitored”. However he was correct saying that they were “self-inflicted and egregious”. Now it matters to secure whether they were “flawed” irrespective of whether the book was in “post mortem” mode already.

As one may surmise, given the well advertized positioning of CIO and well advertized “flaws” in CIO positioning “in hindsight” no later than May 10th 2012, those positions would certainly have never recovered the $6 billion losses in a predictable future right? Well the next part will show that instead, despite all the drumming, all the “off the charts” allegations despising CIO’s positioning, despite all the distortions created by the bank and regulators next, the P&L would have recovered by the end of 2013. To be sure, nothing was left to chance here by the firm after the 12th March 2012: CIO had to lose on every single exposure. All this was elevated and all this went further in loss terms for CIO. This is not possible any more to believe that the firm got surprised at any point in time or even tried to limit those losses snowballing in the “tranche book” of CIO….

c. The positions would have recovered in 2013 in full had CIO kept the book despite their alleged “flaws”. This also was known very well inside the bank. The collapse thus occurred as soon as the projected maximum loss point was reached for skew related exposures at CIO in June 2012.

It is another well accepted misrepresentation that CIO was wrong on its positioning, believed itself so at one stage in March 2012, and thus “doubled down” to try to cover it up. It is a complete mischaracterization of the events on every possible angle. And all regulators would learn it as they scrutinized the emails and slides of Iksil. Still they would ignore the very actions of Iksil in his real job and would conveying a complete misrepresentation of the facts on the matter. CIO did NOT “double down” but secured a positioning that was sensible and that would maximize the likelihood to recover all the losses at a minimum trading cost looking forward. It would have worked against what the seminal articles of April 6th 2012 tried to trigger. It would have worked still despite the misleading statements of Dimon on May 10th 2012. It would have worked quite well onwards even through the restatement, the loud accusations and the blameful settlements of 2013.

113

114

One must accept the view that, if this harsh but highly superficial criticism of CIO’s positioning was to be trusted per say, no recovery was expected before November 2013. And yet….Just to put the coming chapter in context…. Dimon stated that CIO’s positions were “flawed, poor executed, poorly reviewed, poorly monitored”. It was shown before as a matter of fact that the bank HAD compelling reasons to monitor them daily as CIO alone on IG9 10yr if only because CIO was about 150% of the whole market as of January 1st 2012. That was plain impossible that CIO was alone in Jp Morgan, which implies that the “IB universe” was ALSO alone as big as the market going opposite to CIO as of January 1st 2012 already. Thus the CIO trades and positions were necessarily closely scrutinized, and monitored by ICE, by the bank, by the IB and by all the main market players. Second, it has been shown that CIO and bank chiefs ordered to grow some positions in plain awareness of the trading costs. What unsettled their plans, whatever they were, would be the refusal of Iksil to trade further as of March 1st 2012 first time in the morning. The slides and emails of Iksil would only complicate their life further on the matter. Thus here the CIO positions were closely reviewed if only because CFO and CRO had granted temporary but unlimited limit extensions. More, Human Ressources, Drew and Macris had demoted Artajo right after one alert of Iksil of 31st January 2012 and still ordered to keep trading. Artajo had been demoted and had to convince Iksil to trade and trade again. Had Iksil been told that Artajo had been demoted, could this setup have had a chance to work? No. Third it has been shown that the “flaws” were not so obvious as the sneaky “drift” was still very well balanced by other positions in the book until the lethal March 12th 2012 meeting with Ashley Bacon. One may doubt that the positions as such were flawed since this is SOLELY after that date that small but regular disruptions in market prices occurred all at the same time, against every long term significant historical relationships.

The only common denominator, aside from Bacon, was that those disruptions all surgically hurt the CIO “tranche book” alone. The $2 billion losses of late April 2012 were NOT induced by the alleged “flaws”. They were just due to the “magic wand” of this March 12th 2012 meeting with Bacon, the deputy firm-wide CRO who worked on Dimon’s long planned “credit exotics wind down” plan. What were they all waiting for to make this simple, preliminary, planned, seamless, well prepared “CIO to IB” transfer anyway?

That observation alone that JpMorgan top executives delayed at their sole convenience an easy transfer for so long is evidence that the firm certainly did not “escape by a hair” a surprising catastrophe that was in the making at CIO since the start of 2012. This tale has just no reasonable ground as all the former charts showed as well. Still one may argue that CIO may have sort of “pressured” the markets in 2011 or before. That may have been the mistake of CIO, ie a bad execution but that happened in 2011 not the year after. The year 2012 as such may thus have been a sort of “pay back” against the “caveman” trades of CIO.

This part will prove that there was nothing like that. One will observe in particular that there never was any pressure induced by CIO’s trading activity, even in 2010 or 2011 or before. The CIO’s positioning in 2011 and 2012 was sensible, not speculative, focused on long term protections. It was mostly a legacy that Drew would opt to grow rather than let die at the size that it had at the end of 2010. And , despite the huge predictable trading costs that Drew’s orders had induced, her crazy trading strategy would have worked well over time even before the stigma was “settled” by Dimon with a couple of checks totaling officially $13 billion in late 2013.

It matters to first summarize the main features of the “tranche book” as they evolved through the first months of 2012. The figures here are really projective but they are quite close to what an accurate calculation would give based on internal data at Jp Morgan. The first table provides the protection that the book brought against defaults of high profile corporate be they in the IG or in the HY categories:

114

115

Block Early January 2012 (in End of April 2012 (in April 30th without On $milion) $milion) the run risk (in $milion) Itraxx On the run +48 -87 ? Total (off-the-run and +232 +139 +226 on-the-run plus tranches) CDX IG On the run +18 -244 ? Total off-the-run and +252 +72 +316 on-the-run plus tranches Average P&L per +242 +100 +286 default Itraxx and CDX

As one can see in the table above, unless a very unlikely default occurred on the on-the-run index constituted with the most reliable credits in IG, the “tranche book” offered a very decent gain for any default in the IG sector. (to be sure “226=139-(-87)” and “316=72- (-244)”)

The following table will show that, as per Drew’s specific and repeated instructions, CIO had a quite material protection too on HY markets in the US:

CDX HY Item Early Jan 2012 (in $million) End of April 2012(in $million) HY14 to HY17 hedge +127 +245 HY10-HY11 -250 -186 Total without HY8-H9 -123 +59 HY8-HY9 234 +310 Total all-in +111 +369

Thus the “tranche” book of CIO was balanced between a generic protection against default and a net long risk exposure on European spreads. The balance of risks could be projected say in replicating the last crisis to date that occurred between June 2007 and July 2009. The table below provides the order of magnitudes at play in the balance ordered by Drew since the 9th of December 2011:

Item Default P&L per default P&L impact P&L P&L total number (30% recovery) for defaults outright in $ millions OTR Itraxx $46 billion -5 520 on -5 520 outright notional amount 300 bp Main 6 +177 million +1 062 +1 380 euros Xover 20 +48 million euros + 960 +1 248 CDX IG 6 +316 $ million +1 896 +1 896 CDX HY 30 +59$ million +1 770 +1 770 Total US +3 666 +3 666 Total +6 294 protection

The table shows how the gains from a big crisis could be projected to come in the course of April 2012. Other stress scenarios had been communicated in early April 2012 warning that the “tranche book” otherwise could

115

116 lose say $300 to $800 million on top of the current loss. Thus one can see at the time that one could project a $700 million net gain in a major crisis (+6294-5520 as displayed above) and one could expect a further loss of $700 million in an adverse scenario for this “balance of risks” as ordered by Drew (undisclosed here but disclosed many times then). Thus depending upon where the cursor was on the current estimate P&L year to date loss, any manager knew that this figure could range by -$700 million to +$700 million from the 22nd March 2012 onwards (this is the day when the final balance was completed).

These projections showed clearly that there was NO expectation or intent in this “ordered” balance by Drew to forcefully recover the accumulated $1-2 billion loss through a crisis or a definite targeted risk based on markets direction. The long risk simply balanced the huge protection on defaults. That was it: CIO could sit on it expecting P&L swing of “+” $700 million to “-“ $700 million. The P&L “order of magnitude” therefore was $1.4 billion… no less… The awareness was plain through all the hierarchy that the current year to date estimated loss was NOT due to a specific risk or imbalance. It was due to the “drift” that, since March 12th 2012, had spread through just all the positions of the book: the “longs”, the “shorts”, the “long-short”, the “barbells”, the “butterflies” whatever else one can imagine…The expectation at CIO simply was that this drift would “mean revert” ie invert its path one day irreversibly. This is what will happen in 2013 onwards.

How then did the CIO expect to “recover” the current accumulated “estimated Year to date loss” all driven by those weird but timely drifts? The view was that either those drifts would simply mean revert if they were NOT a series of well organized manipulations, or that over time the book would earn a “carry” everyday provided the book was not unwound in any way of form. This “carried” daily gain would simply result from the distortions in prices that the drifts simply had signaled. Yes those drifts had generated arbitrage opportunities where those who were at profits against CIO positions were simply losing their early gains back a little every day that went on….How would it happen in practice? No-one knew but the ultimate outcome was predictable. The drift had caused the curves of the forward spread to steepen to crazy levels. Over time, the CIO position would simply “roll down” those curves and bring a growing positive “carry” little by little. If defaults occurred in the IG9 index the CIO would benefit from front loaded gains and this would force even more those pushing the drift to unwind hurriedly. Thus CIO had a perfectly sensible positioning in this “wait and see” approach. This had definitely NOTHING to do with any “double down” strategy. That was a thing that every investigation team at every regulator would get 100%. Yet they would all convey this very misleading view that CIO had tried to “double down”….The events of November 2012 related to MBIA will prove that CIO was absolutely right on its strategy here to” wait and see” for the defaults to come or the roll down to come. The chart below will plot how the IG9 1 year forward spread curve moved since December 2011 until April 2012….

116

117

220

200

180

160 Fwd as of Dec 2011

140 Fwd as of Jan 2012

120 Fwd as of Feb 2012

100 Fwd as of Mar 2012

80 Fwd as of Apr 2012 60 Proj FwdSpread as of Jan 31st 2012 40 0 2 4 6 8 10 12

As one can see for illustration, the projected forward spread curve “as of January 31st 2012” in orange ( projected for the end of April 2012), using market neutral assumptions, is quite different from what the curve would be at the end of January 2012. There was in February 2012 already the precursory sign of an intense manipulative activity on the parts of some dealers. For the record, one other sign was that the very illiquid IG14 and IG15 indices had outperformed the IG9 in the market rally of this month. One can see that this distortion just kept growing in the following months. The IG9 10yr had actually a risky duration of about 6 then. And one sees that the forward spread in 6 years remained as a sort of “fixed point” in price terms while the rest of the credit planet had rallied. This also is a sign that a longstanding manipulation had been going on in the first 4 months of 2012. All the 1 year forward spreads in 6 years had rallied significantly with the general market rally but ONE, namely the IG9 one related to the IG9 10yr index. This was in summary the opportunity that had grown more and more obvious all along those first 4 months of 2012. This is what was expected to happen soon or later once the manipulation against CIO would ease, provided the book was not unwound.

The next chart will thus add the projected forward spread curve on IG9 so that one can see the kind of opportunity that was:

117

118

220

200

180

160

140

120 Fwd as of Dec 2011 Fwd as of Jan 2012 100 Fwd as of Feb 2012 Fwd as of Mar 2012 80 Fwd as of Apr 2012 Proj FwdSpread as of Jan 31st 2012 60 T Proj FwdSpread as of April 30st 2012 40 0 2 4 6 8 10 12

As one can measure with the arrow, the forward spreads on IG9 based on the price of the IG9 curve ( 5yr, 7yr, and 10yr) were actually underperforming by some 45 bps so far from January 1st 2012 till April 30th 2012. The basis point value was in the range of $30-40 million which was somewhat worth between $1.35 billion and $1.8 billion in total. The loss ‘year to date” was about $2.1 billion. Thus, as explained, this would not explain by far the total loss, but that shows that, even with the effect of the March 12th 2012 meeting with Bacon, CIO simply could rely on a “mean reversion” for the sole IG9 forward curve to recover most of the loss accumulated so far. The same phenomenon indeed was observed on the Itraxx Main S9 forward spreads. Here the basis point value was about $15-20 million and the cumulated drift was in the same magnitudes. Thus on top of the recovery of the IG9 curve CIO could rely on the simultaneous recovery of the Itraxx S9 curve, something which would have brought a gain of $700 to $900 million likely so. Adding the 2 together, CIO expected a gain ranging between $2.05 billion and $2.7 billion. If that did not come because market resumed a normal behavior on the forward spreads, this would have come in the form of a positive “carry” day after day. All the players going against CIO knew that pattern very well. The bank chiefs knew it as well, like Ashley Bacon. All the “secondary” losses that the March 12th 2012 meeting had brought and that the seminal articles had just made worse would be recovered.

CIO had enough rational arguments to NOT be forced to unwind despite the monstrous loss in late April 2012. And more CIO had just to wait, unlike Dimon, unlike the IB, unlike Bacon, unlike the hedge funds, unlike the dealers all teaming up here against CIO. This was written in the emails and documented in detail. What about the agenda of the regulators who had sent so many warnings in late 2010 about the CIO? They were waiting already but they would NOT try once to meet with Iksil before July 2012…Later on, through their investigation they shall all try to push the misleading tale that CIO was in a rush to try to recover the losses and/or was in denial of the loss magnitude itself.

One wonders why Dimon was still postponing this so simple transfer of the “tranche book” of CIO towards the IB as of April 30th 2012….He would be rumored to want to “see all the positions. Now!” then, ie by the 30th April 2012. He had already dispatched Bacon and O’Rehilly, his avowed trusted lieutenant to take over the “tranche book” at CIO. Nothing therefore stopped him to simply order this seamless, preliminary, planned, well prepared easy transfer from “CIO to IB”…..He did not though make the move….He would weigh his

118

119 communication options instead as Drew would write it to herself on the fully traceable email system of the firm in CAPITAL letters per May 3rd 2012….

Now the issue was that no one in the markets could attempt to exit that situation. There was a need to secure the loss at CIO without showing yet to the market players that Jp Morgan anyway would NOT unwind anything in the markets for what would be just the “tranche book” transfer CIO to the IB in the end. The hedge funds and dealers going against CIO so deliberately as the media reported sat on massive gains tantamount to the CIO losses. But the embedded costs to exit their trades as much as the CIO ones were enormous. They may lose faith and start speculating with good common sense whether they had not better start getting out before JpMorgan made any move. They knew it…The “tranche book” had been too big since 2007. Therefore only an internal collapse was the ultimate solution for Jp Morgan actually. But for that to happen this meant forcing Drew and CIO and therefore depositors to accept a massive loss that would profit “traders”, “executives” and shareholders. That may spark a pretty bad headline news for them the manipulators. They had enough reason to worry about their accumulated gains….

It has been shown in the first part of this document that the bank had big stakes at play here too. The dilemma probably was: the bigger the loss for CIO, the bigger the gain for the bank, but the bigger the reluctance of CIO to accept this loss, and the more shocking it would show on the headline news. The “London whale” scandal was here to stay in early May 2012…And the storyboard did not look good…The issue involved regulators here: CIO was meant to protect depositors through its “wise” investments of “strategic liquidity reserve”. The regulators watched it since 2006 quite closely. The regulators were officially quite concerned since 2009. A plan to “sell” the “CIO tranche book” to “credit hybrids” at the IB had probably failed in late 2009. The book had to be killed in early 2010. This was NOT possible since markets were illiquid. The regulators demanded a drastic plan for a “CIO clean up” since early 2011.They would put their running investigation “on hold” in February 2011 conveniently so. CIO was meant to be broken up actually since 2011 as Artajo would confirm. This of course included the sure death of the “tranche book” of CIO. Regulators were waiting still while the bank itself may have wholeheartedly made this simple preliminary transfer of the “tranche book” from CIO to the IB in late 2010…But the bank had not done that. And no regulator had requested that yet….That looked quite bad…

Thus in early May 2012 the bank still had to finish the “job” of stuffing the losses in this CIO book that was about to die imminently anyway. What was the goal? Anchoring the market manipulators in their initial expectations the time for the IG9 10yr skew to get to zero. It had to look like an egregious error of the CIO staff, of Iksil in particular, so that it could never be interpreted as a diversion of “depositors interests” in favor of “traders”. Thus the litigations would be avoided on matters regarding, market manipulation, depositors money diversion, media manipulations, mismarking due to absence of liquidity reserves since 2007…..This is the context of May 10th 2012, of the 10-Q the statements of Dimon and of the ensuing investigations.

The table below gives a very useful view of the issue at stake that existed since 2007 internally at Jp Morgan. The figures are just recollections and cannot really be very wrong. But they may not be 100% accurate. Again, 5 years on, the bank could disclose publicly the exact inventory of positions at CIO and people could check…. If really all this is “behind us” at JpMorgan…. The table below provides just a proxy for the outstanding amounts at CIO of the main exposures for the end of 2011 and for the end of April 30th 2012. The table shows for each key exposure, the bid-offer and the P&L impact of a simple unwind ie a standard trading size as accepted by a dealer. One has to bear in mind here that, the liquidity was so poor that the dealers would shift their prices adversely against the client’s interest as soon as they would have traded. The shift itself would be a 3rd or half the bid offer spread typically. For the IG9 10yr for example they quoted in a 2-3bps range and would shift their bid and their offer by 1 bps. Now the exposures at stakes ranged between $40-50 billion (at least) at the IB in one way and $81 Billion at CIO in the other way, while dealers would only trade by one clip of $300 million.

119

120

This meant that the operation to unwind all this involved at least some 150 trades and 1bps shifts for each trade in theory..... That was scary indeed for anyone. Only the internal collapse of JP Morgan IB with Jp Morgan CIO was reasonable in fact and all the market players had been made aware of that by Jp morgan since late 2011.

One will see that the total exit cost of CIO positions was already, for one standalone standard trade size, of $370 million as of January 2012 the 1st. Drew , CFO, risk management and all the other executives at CIO had been alerted of that by Iksil, Artajo and Grout all along 2011. This is what had already led in June 2011 to plan a “run –off” inducing the temporary creation of “Strategy 27”. Now as of late April 2012, the unwind cost had shot up knowingly so from $370 million at the end of December 2011, to more than $1.25 billion. The unwind costs had basically been multiplied by 3 or more. The order of magnitude is telling also of the fact that a $5 billion loss was easily reached just because the prices would shift by 3-4 bid-offer ranges in “IG equivalent” terms, which was quite frequent in the course of a quarter or 2. Drew here had ordered to a massive increase in full knowledge witness her anger when Iksil by the 18th January 2012 told her on a conference call that the unwind costs had grown by about 40% between the 31st December 2011 and the 17th January 2012 (yes only 2 trading weeks!)…..

Thus the table below is very instructive. It was a very basic analysis that risk management performed regularly, on top of other much more sophisticated studies on the book. Keith Stephan and Evan Kalimtgis were experts in this field. This did NOT require much modeling or “help” from traders. It was quite explanatory and intuitive for anyone to see inside Jp Morgan. As explained before, for the sake of comparison, the amounts below are displayed in their outstanding net total amounts. The “P&L impact” provides the cost induced by the “bid- offer”. One might be surprised to see that the IG9 block trading costs were NOT a big share as such of the totals for the “P&L impact”. This is in part what justified the quite misleading “explanation” that Drew wrote for herself in CAPITAL LETTERS in early May 2012 that the IG9 was the best “proxy” in liquidity terms. This also by the same token emphasizes the manipulative side of this “45 bps” cumulated underperformance of IG9 forward spreads. Indeed one struggles even more to understand why then the IG9 curve underperformed so much even the most il-liquid other IG curves at the time when the markets rallied. Thus one can see that the IG9 “block” was NOT the only one risk in the book actually by far, but was even more clearly manipulated in 2012 (how can one manipulate a more liquid market versus a much less liquid market like for example IG9 versus IG14 or IG15?):

Item Dec 2011 Bid-Offer P&L April 2012 Bid-Offer P&L Inventory Value* Impact Inventory Value Impact Outright Main -22 000 1 10 +46 000 2 41 S16 HY17 vs Main + 9 000 2 8 + 5 000 3 7 S16 Xover Vs Main + 10 000 2 9 +10 000 to 3 17 S16 +15 000 HY17 vs IG17 + 5 000 1.5 3 +80 000 3 107 IG14-IG15-IG16 -40 000 3 56 -40 000 5 88 vs IG17 Main S9 Fwd vs Main16&Xover16 Index Forward +10 000 3 & 3 27 +25 000 5&6 83 0-3 5yr ( x 15) - 2 500 1% 24 -3 000 2.3% 65 0-3 10yr (x 3) + 1 500 1% 14 +1 000 1.7% 17 IG9 fwd vs IG17&HY17 Index forward +51 000 1 & 2 39 +81 000 4&5 198

120

121

0-3 5yr ( x 20) - 2 500 1.5% 18 -3 500 2% 68 0-3 10yr (x 4) + 3 000 1% 17 +2 500 1.5% 37 HY14-…-HY16 -10 000 0.4% 52 -20 000 1% 199 vs HY17 HY10-HY11 vs + 5 000 0.4% 23 + 5 000 1.5% 84 HY17 HY10 10-15 5yr + 1 000 0.5% 5 + 1 000 1.5% 13 HY10 10-15 7yr +3 000 0.5% 14 +4 000 2% 85 HY11 10-15 5yr + 1 000 0.5% 6 +1 500 1.5% 17 HY8 10-15 5yr -1 000 0.5% 5 -2 000 0.5% 8 HY9 10-15 5yr - 500 0.5% 3 - -200 1% 2 Super-Senior S9 +15 000 2 14 +35 000 5 78 Super-Senior IG9 +20 000 2 19 +30 000 4 48 Total** +350 000 $368 +600 000 $1 262 ** the total simply sums up the notional amounts to arrive at a rounded number. This is provided for the sake of intuition. Once again the IG9 block was big, the bigger one of 3 to 4 other similar blocks that were of the same order to magnitude anyway.

Looking at the table here, one can get quite easily an intuition of the risks that were in the “tranche book”. Indeed as the prices moved in proportion and as the bid-offers reflected the risks of the instrument being quoted, the “P&L impact” gives a direct measure of this “unwind” risk. On the table above, circled are the main risks, other than the tranches related risks. One can see that the IG9 forward risk was about $198 million. But one can see that the “decompression” trade was similar in size looking at the lines “HY14-15-16 vs IG17” and the line “HY17 vs IG17”. Another similar risk was found in a pure relative value trade between HY10-HY11-etc.. versus the benchmark HY17. Likewise at the bottom of the table, the “super-senior” tranche risk was of a similar magnitude. The HY tranches were also of the same order of risk. The same comment can be made about the Itraxx Main S9 tranches combined with the IG9 tranches. Thus if one summarizes the situation, be that for January 1st 2012 or April 30th 2012, the IG9 10yr related risk weighed about one 6th of the “unwind” risks of the CIO “tranche book”. There were 5 other risk buckets that were also quite well known, namely: the “decompression” risk, the “HY off the run versus HY on-the-run” risk, the “super-senior risk”, the “HY tranche risk”, and the “IG9 and Main S9 tranche” risk. They were all of the same order of magnitude. Thus, fundamentally the IG9 10yr risk was just 15% of the total exposure of CIO when markets function normally. But here they were NOT functioning normally, obviously so.

Thus as of May 10th 2012, all was clear. The unwind costs were the key. There was no need for a massive market shift against CIO to record multi-billion losses. This adverse market move just had to be done surgically against every single position of CIO. And this is precisely what started happening right after the March 12th 2012 meeting with Ashley Bacon and Ian Green whereby CIO officially had surrendered to IB. The “transfer” was ready to go as far as Drew was concerned since Marc 23rd 2012. CIO was prepared to surrender the ’tranche book’ in a transfer to the IB as planned no later than the 26th Match 2012, any day at Dimon’s entire convenience. That was the sole possible “wind down” strategy envisioned at JpMorgan anyway.

Strangely enough the focus of the markets was on IG9 10yr index while this was only 15% of the “CIO tranche book”. The CIO was not alone in the “jpmorgan” equation for sure. Drew had ordered a strategy that was quite intuitively dangerous and paradoxical in the context where both CIO and its “tranche book” were to be dismantled since 2011. The March 12th 2012 meeting with Ashley Bacon, called by Macris as of March 1st 2012, had started a surgical spreading of losses in the book that only came from “drifts” showing across just ALL the positions of CIO specifically. This loss did not require large price shifts IF they were well synchronized as it turned out to be the case. The April 6th 2012 seminal and manipulative articles had hammered the IG9 and Main

121

122

S9 curves further mostly. Ironically this entitled CIO to claim that, despite the quite remarkable surgical snowballing of losses since Bacon was involved, it just had to wait and was NOT forced to unwind be that with the IB or with the markets…. “Transfer now or not transfer now?”... Here was the Gordian Knot for JpMorgan: CIO had to capitulate internally but CIO could wait….. Thus, as Drew wrote to herself in early May 2012, in CAPITAL LETTER on her bank email for everyone to see later, Dimon and Braunstein had weighed their communication strategy in “risk//reward” terms.

The checkmate about the liquidity issue and unwinding other than internally was crystal clear. The loss at CIO was surely a gain for the bank as the start of this document showed it. But it had to be “locked appropriately” for all the stake holders, namely the bank, the regulators and the market manipulators. The “risks” and “rewards” were thus well known in this highly manipulative context (media, markets, accounting, emails, meetings, holiday, reported 10-Q etc…). Dimon thus issued the 10-Q, made statements while Alllistair Webster issued both confidentially the new firm valuation policy and his audit memo of the CIO London valuation for March 30th 2012. As it was said on the website, a peculiar mismarking was created out of the blue here in the middle of all these publications and disclosures. Clearly this well manufactured “mismarking” would force CIO out of its positions before the H-hour projected by the CIO front office staff who was NOT in the secrets… But would it work fast and well for the others? A couple of inflammatory statements against CIO would help no doubt. What was the impact in the market and other “collateral costs” of this Dimon’s carefully crafted communication strategy of May 10th 2012? Drew would leave but Iksil could NOT be publicly blamed by the CEO. However the CEO will completely distort through his characterizations the facts concerning Iksil. That too was carefully crafted although very gross….That was a risk looking forward if the “French trader” spoke up one day.

The coming chart will display the “reward” in the most comprehensive and intuitive manner. In order to avoid any debate about the modeling of this complex book and positions, some intuitive price moves will be provided. There will be the general market price level given by the on-the-run index. There will the IG9 5-10 price difference that reflected relatively well the loss on the forward spreads be that for IG9 or for Itraxx Main S9. There will be also the relative value of the very, very, very illiquid IG15 index versus the most liquid IG17 index for the 5 yr and for the 10yr. If the IG17 spread moved up, CIO lost. If the IG15 5yr outperformed the IG17 5yr, CIO lost. If the IG15 10yr underperformed the IG17 10yr now, CIO lost too. If the IG9 5-10 spread moved up, CIO lost a fortune. The chart that comes provides a long term history that is a recollection that should be quite good. It spans from the end of 2010 and extends to 2013. The prices before 2013 are reconstructions. The prices post may 2012 result from a regular monitoring of the prices being made available on the web on a daily basis. Once again, the data exist and should be publicly disclosed for anyone to check the chart below:

122

123

2,20% 1,40%

1,30% 1,95%

1,20%

1,70% 1,10%

1,45% 1,00%

0,90% 1,20%

IG9 5x10 0,80% IG15xIG17 0,95% IG 17x15 10yr 0,70%

OTR 5yr IG

0,70% 0,60%

16/02/2011 18/11/2010 18/12/2010 17/01/2011 18/03/2011 17/04/2011 17/05/2011 16/06/2011 16/07/2011 15/08/2011 14/09/2011 14/10/2011 13/11/2011 13/12/2011 12/01/2012 11/02/2012 12/03/2012 11/04/2012 11/05/2012 10/06/2012 10/07/2012 09/08/2012 08/09/2012 08/10/2012 07/11/2012 07/12/2012 06/01/2013 05/02/2013

As one can observe on this “complex” chart displaying “complex” prices that would impact this “complex” firm, Dimon’s statements of May 10th 2012 caused in a matter of 2 weeks a crystal clear distortion in prices against CIO’s positions that had a crystal clear magnitude similar to the whole European crisis of 2011. The difference between the 2 periods is clear: the first jump is due to a crisis in Europe that hit the whole planet, while the second jump was the result of the carefully weighed strategy of communication of Jp Morgan on May 10th 2012…. This comparison leaves the March 12th meeting and the seminal articles in oblivion with regards to the magnitude that was achieved here in the price distortion. They are the ones that led fast to the $6 billion loss at CIO 2 weeks thereafter as some articles reported in June 2012.

This chart above therefore erases in full the relation anyway between the alleged “price pressure” from CIO and the resulting $6 billion loss. All this occurred solely due to a carefully crafted blame on CIO by Dimon on May 10th 2012….

Thus the $6 billion loss would be almost entirely caused by the statements of Dimon on May 10th 2012 regarding the “tranche book” of CIO. $4 billion will pop due to these statements in a matter of 3 or 4 weeks no more. Was it really a mistake? Remember here what Drew would write in CAPITAL letters to herself in an “NOT TO BE MISSED” email dated May 3rd 2012: risks and rewards had been carefully weighed here. There were $60 billion in tangible capital gains as planned versus a $6 billion loss that was actually securing these $60 billion tangible capital gains. That was A reward indeed. What was the “risk”? Well the litigations and investigations may lead to serious accusations of manipulation if the executives were made accountable on a personal basis. Iksil was not yet thrown officially under the bus. But he would be and the risk was that he could

123

124 speak up one day. The risk never materialized as one knows today until mid 2017. The communication job would thus be “well done” on balance… The long prepared “script” leading to the may 10th 2012 “message” of Dimon is therefore worth a closer look.

In 2011, through the European crisis, it was a general market shift that was very broad and fueled by deep economic concerns about Europe. Here CIO had made little money on the “tranche” book (about $150-100 million) while CIO lost many billions in its $360 billion investment portfolios. The orders of Drew came in December 2011 to shift the profile of the tranche book, thereby locking some of these small gains and lining the “tranche book” risks more towards the existing $360 billion of investments that bled $billions already. Was it wise, liquid and conservative? No, it amounted genuinely to a “double down” made not on the “tranche book” but on the total CIO. Was it noticed by the regulators then? Yes it was. And in May 10th 20112 onwards, the critical shift in CIO losses was only sparked by Dimon’s statements (many of which were misleading especially the “flawed, poorly monitored, poorly executed, poorly reviewed etc…”). Then the CIO lost about $2 to $2.5 billion in the following 2 weeks before the end of May 2012…..And another $2 billion more in the 2 weeks thereafter…. If one wants to summarize the difference one has only to look at the line “IG15xIG17”. In 2011, through a real crisis the 5yr IG15 underperformed the more liquid IG17. In 2012, it outperformed which is a clear indication that markets were manipulated in 2012. Dimon’s statements of May 10th 2012 only made this manipulation take a more massive dimension in a very short while. Nothing therefore was “out of control” as far as the loss at CIO was concerned. Instead it would made much, much, much, much worse and on purpose…

But one wonders what the hell this purpose here could be?! It matters to spend a little time on this IG15 index which outperformed the IG17 between mid November 2011 and December 2012 right when the markets had rallied quite substantially. As one could guess the CIO “tranche book” had a short risk exposure on the IG15 and therefore lost a lot of money. Here one of the least frequently traded index, namely the IG15 (did it ever trade in the period actually?) saw its spread tightening in proportion much more than the one of the most frequently traded index, namely the IG17. The IG15 was so notoriously illiquid that dealers would not even offer a tentative quote in 2011 or 2012. If they did it, that was just “for indicative purpose”. If one wanted to trade, that was NOT possible in fact. The “client” had to tell the “dealer” his target price and the dealers would do his best with no guarantee of success. If by chance a trade was executed, the size would be small like $100 million (to be compared with the meager $300 million clips for the IG9 10yr). As a comparison the IG17 5yr would trade by clips of $500 million to $1 billion every 5 minutes at its highest liquidity point. This was the case in the first quarter of 2012, not every day for sure but often enough, when all the credit spreads moved tighter by 20% to 45% or more.

The question then is: “how could the IG15 spread move more than the IG17 without trading while the IG17 traded so frequently?” The pattern was noticeable in the fast rally of January 2012. Iksil had flagged it to his management pointing out here a clue indicating some manipulation of prices. Indeed, only the dealers, not trading the IG15 and refusing to trade the IG15 even when they quoted it, could have artificially lowered its quoted spreads faster than the benchmark index, ie the IG17. Why would they do that? Was it because they had a position for themselves or their clients on the IG15? No they had no position since they wanted to have none. The IG15 market was plain dead since mid 2011: everyone had shifted from the IG 15 into the IG16 and next into the IG17. The only reason left was that they were tipped that the legacy positions of CIO on the IG15 would soon be eliminated. Who other than JpMorgan itself could have done that leak? That one leak had occurred in early January 2012, not later. And in this IG15 plain dead market that no dealer would like to trade anyway, only a collapse done internally at Jp Morgan would be possibly a credible leak to make the dealers confident in their manipulation of the IG15 quotes. Thus they were tipped, not by CIO, but by Jp Morgan that this collapse was to happen. The dealers did the rest of the reasoning if only because they were just speculating…

124

125

How would they be convinced that it was in their own interest to move the prices like this? It likely was explained to them that it was in their interest that the IG15 index should trade much tighter than were it was in mid November 2011. They were told that CIO likely would fight the IB here. Dealers had better put themselves off any attractive levels for CIO to unwind with them. Otherwise they would have next to turn towards the IB to unwind this IG15 “dead” index trade. Better was to leave CIO in front of the IB all along and avoid the JpMorgan family affairs. They should lower their quotes as much as they could - thus without trading - merely to dissuade CIO to trade with them…. The dealers would comply willingly to some extent. What would be then their own rationale for that quite un-natural outperformance? Well the only plausible reason is to be found in the skew and the single name constituents. Here the IG9 10yr skew “believers” had second thoughts. The IG15 was containing very high quality constituents on a single name basis. But, since the index was very, very illiquid, it traded very wide relative to the more liquid indices. The IG15 index shared like 60-70% of its constituents with the IG9 index. And the IG15 skew was potentially “movable” (for a limited period provide the index did not trade indeed) to very small numbers in implied skew since the index was as illiquid as its constituents were anyway. Thus, in order to engineer a tightening of the IG9 10yr skew, which was the cornerstone of the coming $50-60 billion of gains in tangible capital for the bank, one had to move the IG15 much, much tighter. Then it would force the constituents to move alongside the IG15 index, without trading anyway. Thus this would make the coming IG9 10yr skew tightening down to zero “credible” in that the manipulation was less visible than if it had hit the IG9 10yr alone.

Here Jp morgan had a potential “business” argument for dealers like CITIGROUP, CSFB who all had inherited massive skew exposures. Accepting the tip from Jp Morgan and moving in concert the IG15 artificially tighter like this, they would improve their own financial performance. This weird but quite visible pattern started in early January 2012. The carefully crafted statements of Dimon on May 10th 2012, weighing “risk and rewards” in Drew’s CAPITAL letters, would operate the same effect but with an historical magnitude this time. Once again, the data exist and it would be easy to check the skew exposures of the main dealers mentioned above, the role played by the IG15 outperformance in facilitating the long planned collapse of the IG9 10yr skew through the “London whale” scandal to zero.

There would be then NO doubt as to whether the bank was just lucky to have made $50-60 billion through the event. It had been somewhat “lucky” or rather was actually quite actively preparing all this setup no later than mid November 2011. Those statements of Dimon of May 10th 2012, as mentioned made the year to date loss at CIO skyrocket from $2.5 billion as of say May 9th 2012 to more than $4.5 billion as of say May 16th 2012. The total number in the beginning of June 2012 would reach $5.5 to 6 billion already…..Thus one can state that the May 10th 2012 statements of Dimon actually made the $6 billion loss at CIO for at least 2 thirds. As to the first third that spread through the first 4 months of 2012, if the move of the IG15 is any indicator, it also came from statements of Jp Morgan staff but in the form of sustained rumors about CIO collapsing its positions soon and winding down some IG15 exposure on the way. One could understand much better how Weinstein could posture in the very offices of some Jp Morgan headquarters in New York as of February 2nd 2012, and vaunting the appeal of IG9 10yr skew trades targeting “zero”. One could better understand in what sense some journalists at Bloomberg saw in Weinstein a truly “legendary” investor.

As was said before, the risks of the “tranche book” of CIO were balanced in many ways. The CIO could afford to simply wait and would recover the loss. The issue for CIO was well flagged since 2011 though: few bid-offer changes in prices would induce massive losses in the meantime. Any market unwind of size was precluded since late 2010. One thus really struggles to understand the stubborn posture of Drew with her orders starting in December 2011. What did she elevate to HER boss all along? He reportedly gave her a bear hug on May 14th 2012 when she was retired in haste. How reliable all this was? No-one however had ordered yet this quite

125

126 simple and planned preliminary “transfer of CIO to the IB” for the “tranche book”. The top bank executives could have launched this transfer anyway as this did NOT force CIO to accept this ‘$300million’ difference as “CIO’s loss” but it froze the prices as they were, for all regulators and other third parties to see…..Maybe the latter did not want to see those prices yet.

Now one wonders: “What would have happened next, after the restatement in July-August 2012, if the “tranche book” positions had not been “off-shored”? The following chart sketches how the IG9 forward spread itself performed versus the IG market.

0,45% IG9 fwd Bp diff vs Pure IG

0,40%

0,35%

0,30%

0,25%

0,20%

0,15%

0,10%

0,05%

0,00%

03/11/2010 03/12/2010 02/01/2011 01/02/2011 03/03/2011 02/04/2011 02/05/2011 01/06/2011 01/07/2011 31/07/2011 30/08/2011 29/09/2011 29/10/2011 28/11/2011 28/12/2011 27/01/2012 26/02/2012 27/03/2012 26/04/2012 26/05/2012 25/06/2012 25/07/2012 24/08/2012 23/09/2012 23/10/2012 22/11/2012 22/12/2012 21/01/2013 20/02/2013 -0,05%

-0,10%

Unsurprisingly the savage drift that started soon after Macris had called Bacon on March 1st 2012, stopped around the end of May 2012. From mid June 2012 onwards, this was next the status quo until 2013 with the “London Whale” scandal taking a new dimension. See the “drop” around June 20th 2012 right when the market learn from JpMorgan that the bank has “got rid” of the “tranche book” while nothing got unwound in the street….The “reward” was here for Dimon but the “risks” only started materializing as it seemed…. One can notice again that the carefully crafted statements of Dimon on May 10th 2012 made for the IG9 forward spreads as much damage as the combined effect of March 2012 moves plus the fury caused by the seminal articles plus the fake “collateral dispute”, plus the so-called “I want to see all the positions, Now!”. Those figures are just indications. But it would be quite easy to check. One can ALSO notice that the alleged “massive trades” of CIO actually barely moved the needle in price terms actually between November 1st 2011 and March 1st 2012.

As one can see the forward spreads of IG9 remained in a relatively narrow range that had been the same from October 2011 actually till end of February 2012. To be sure, CIO almost did NOT trade the IG9 from October 2011 to January 1st 2012, ie 3 months. Then CIO traded “massively” in January and February 2012…. No change in the long-term relative price historical range. Yet, as explained before, a sneaky “drift “ had been detected by Iksil and elevated all the way up to Drew. As one can infer too when paying a closer attention to the label “Pure IG”, that this “drift” started relative to the “HY” part that was in the IG9 and also in other IG

126

127 indices as well. One may argue once again that CIO sort of “blocked” a trend in the making about “pure IG” in IG9 versus “pure IG” in peer indices….Or one may argue that this is what the dealers tried to manipulate in vain to collapse the “pure IG” skews across the board BECAUSE CIO was trading….. But if the IG15 move is any guide here, this trend is just a pure manipulation on the parts of the dealers involved in quoting the IG15… CIO had no part since CIO could NOT trade the IG15 then when it tried to noting this crazy outperformance of the IG15. So that IG15 case shows without any doubt who the manipulators were…They were NOT at CIO.

Had CIO then put an undue pressure on the IG15? The call of Gabriel Roberts on March 1st 2012 to Iksil is further evidence of this ongoing manipulation made by the dealers involved, not CIO. Now here one must see that, had CIO kept the trades until 2013 it would likely have remained at a deep loss for a while before the recovery would occur. And the recovery was quite predictable for the positioning of CIO was NOT flawed at all.

How long would it have taken for CIO to recover the $6.3 billion loss that was stated as such by braunstein in October 2012?

Now the coming charts are reconstructions too. The summarized position of the “tranche book” as of end of March 2012 was simulated using prices based on memory for the period before 2012 and based on continuous monitoring post May 2012. The chart below will show an interesting perspective. It does NOT display what the actual performance had been or would have been of the actual book of CIO. It simulated at best how, had the CIO tranche book been all along what it was at the end of March 2012 as per Drew’s orders, this position would have performed since 2010 and until late 2013.

Total P&L with March-April 2012 balance 2 000 Total P&L "risk and reward"... first the 'london 1 500 whale' tale ... Next ' sorry we made a mistake'... which was not a 1 000 mistake....The P&L recovers mechanically over time.... 500

04/10/2011 28/10/2012 07/01/2011 06/02/2011 08/03/2011 07/04/2011 07/05/2011 06/06/2011 06/07/2011 05/08/2011 04/09/2011 03/11/2011 03/12/2011 02/01/2012 01/02/2012 02/03/2012 01/04/2012 01/05/2012 31/05/2012 30/06/2012 30/07/2012 29/08/2012 28/09/2012 27/11/2012 27/12/2012 26/01/2013 25/02/2013 27/03/2013 26/04/2013 26/05/2013 25/06/2013 25/07/2013 24/08/2013 23/09/2013

-500

-1 000

-1 500 Out of Dimon statemens the book was balanced indeed. See the P&L path before the -2 000 'london whale'tale.... It ranges in a +/- $500 million range or a bid-offer value -2 500 See, what the P&L does after June 2012 gestures and manipulations... same range.. next -3 000 the carry works no matter what!

As evidenced in this chart, the “tranche book” of CIO would have sailed through the European crisis of 2011 unharmed, absent any manipulative drift as the ones experienced in 2012. There is NO mystery in that achievement. Iksil had based his analysis for CIO top chiefs on the past year of 2011 to refine the “balancing of

127

128 risks” that “Jamie”, Drew and Macris wanted. This is certainly far from being a “speculation” on the future. This is surely NOT a perfect “balancing” but was in NO WAY a “double down” as far as the “tranche book” was concerned. This is very visible for the period covering January to November 2011. Now one sees a spike in Performance that is simply the effect, including though the early manipulative drifts elevated by Iksil then, of the ECB LTRO and other good news that sparked a rally in the beginning of 2012. Here the net long risk exposure on Itraxx Main on the run indices brought a nice gain. One sees next that the “initiative of Macris” to call Bacon in the collapse process marks the “top” around March 1st 2012 and the crash. Macris had “undone” what he had ordered Iksil to achieve right before. This will be a plunge in a very peculiar context. Weinstein pressures visibly prices in the market at February month end New York close. Roberts wonders loudly what is going on. Drew and Dimon have just run the CIO Business Review. Kalimtgis resigns abruptly. Something is wrong for him, like very wrong and this is NOT about the balance of risks in the “tranche book”. Next there will be the March 5th & 6th dodgy order from New York, the toxic March 12th 2012 meeting where Bacon takes over on the long planned collapse of the “tranche book of CIO”. There will be the seminal misleading April 6th 20112 article, the fake “collateral dispute”. But, all this only brought the net P&L back to zero actually when compared to the January 7th 2011 start….

It is really the May 10th 2012 statement of Dimon that put this positioning at a $2 to $2.5 billion loss. Next one sees that the recovery begins in late 2012 very slowly but surely so! One should notice that the maximum shift “Top to bottom” is only of $4 billion in total. This is mostly due to the fact that this is just a simulation and that the very high carry plays a significant role as early as January 2012 in fact as one sees in 2013. The next chart displays the amounts at stake on monthly carry.

total monthly carry 120 total monthly carry 100

18/03/2011 08/10/2012 18/11/2010 18/12/2010 17/01/2011 16/02/2011 17/04/2011 17/05/2011 16/06/2011 16/07/2011 15/08/2011 14/09/2011 14/10/2011 13/11/2011 13/12/2011 12/01/2012 11/02/2012 12/03/2012 11/04/2012 11/05/2012 10/06/2012 10/07/2012 09/08/2012 08/09/2012 07/11/2012 07/12/2012 06/01/2013 05/02/2013 07/03/2013 06/04/2013 06/05/2013 05/06/2013 05/07/2013 04/08/2013 03/09/2013 03/10/2013 02/11/2013

-20

-40

-60

-80

In this cumulated carry increase starting in early 2012 there was NO surprise. All this was planned and expected by just all the market players who faced CIO’s positions on the “tranche book”.

128

129

Now to have the full picture in sight, the next chart displays the effect of carry. That shows still that this is just a simulation but it conveys at least the right orders of magnitudes:

2 000

1 500

1 000

500

14/10/2011 06/05/2013 04/08/2013 18/11/2010 18/12/2010 17/01/2011 16/02/2011 18/03/2011 17/04/2011 17/05/2011 16/06/2011 16/07/2011 15/08/2011 14/09/2011 13/11/2011 13/12/2011 12/01/2012 11/02/2012 12/03/2012 11/04/2012 11/05/2012 10/06/2012 10/07/2012 09/08/2012 08/09/2012 08/10/2012 07/11/2012 07/12/2012 06/01/2013 05/02/2013 07/03/2013 06/04/2013 05/06/2013 05/07/2013 03/09/2013 03/10/2013 02/11/2013

-500

-1 000

-1 500 Total P&L total carry -2 000 Total P&L

-2 500

-3 000

Such was the expectation of CIO when the executives of Jp Morgan and regulators asked question since early April 2012 while the “post mortem” was acted, while the seminal misleading article had gone out in the media. The book indeed was balanced and would recover the losses nevertheless over time. The book indeed would recover anyway as the orders of magnitude matched. The only issue as can be seen was the order of magnitude itself of the “one touch” bigger loss on the way to recovery since one counted in $billions anyway. Was it so shocking for Jp Morgan’s size?

Well the time horizon was key here but based upon quite a different agenda related to the long planned $50-60 billion planned “tangible gain”. How long this CIO of this big US bank had to wait? The table that comes will show it. The answer was March 2013, ie 9 months. If the “Russia-LTCM” crisis, the “Dot.com” bubble, the “subprime crisis” had been any guide, one could expect a horizon ranging between 12 and 18 months depending on the severity of the crisis. Thus those 9 months were quite long but not extreme. Still this shows that the May 10th 2012 statements of Dimon made quite a commotion in the markets that still compared to significant economic crisis. This as such is remarkable to say the least.

129

130

80,0 The mean reversion shows up at last! 60,0 regime shift in summer 2011 : what happened? Large financial institutions become exposed to the financing issues of G7 governments. This spreads from MBIA to BNP, SOC GEN and Jp

40,0

20,0

0,0

18/11/2010 17/01/2011 18/03/2011 17/05/2011 16/07/2011 14/09/2011 13/11/2011 12/01/2012 12/03/2012 11/05/2012 10/07/2012 08/09/2012 07/11/2012 06/01/2013 07/03/2013 06/05/2013 05/07/2013 03/09/2013

-20,0

The IG9 fwd loses 60 Bps for an average value of -40,0 130 Bps. The Main S9 loses some 90 Bps for an IG9 vs IG17 average value of 160 bps. IN all these 'noise' and S9 vs S16 alleged disruptions, the forwards display still a very high level of correlation! -60,0

The table above allows to compare once again the actual magnitude of the European crisis of 2011 with the distortion that the “London whale” scandal induced. This is massive as far as the IG9 forward spreads were concerned. Again this was just a manipulation of prices done on purpose on targeted positions that happened to be all located in the “tranche book” of CIO then. Could it ever be a coincidence or even “market noise”?

The answers lie in the very weird pattern that the IG9 skew witnessed through the “London whale” scandal. A few charts will finish showing what actually was to happen as planned within Jp Morgan since Late 2011. The following charts are sketching what happened in the first half of 2012. In order to show the distortion on the IG9 specifically the charts will plot since 2007 until June 2012 how the IG9 on a constant 5yr maturity traded versus its components based on the rolling current 5yr CDS of those components. To be sure, for the experts, the skew on the IG9 rolling-5yr will be expressed in basis points. These skew basis points are inferred from the difference in present value of the risky cash-flows coming from the skew position itself that is next divided by the risky duration of the IG9 “rolling-5yr” ie a tenor that is held constant over the years. Thus the spreads will NOT be a simple and misleading difference of running average spreads but fully discounted series of risky cash-flows. To be sure the chart does not make the typical error of displaying a running spread difference between the components’ average spread and the index spread. The chart is based on a present value that next is expressed in basis points using the index risky duration. Thus the charts will respect the way the dealers themselves quoted the IG9 skew as a “package” since 2008 at least. This technical detail may sound weird to the people unfamiliar with the concept of the “risky duration” but it mattered a lot during the crisis. The only thing to keep in mind here is that the charts will show the IG9 skew as the dealers should have quoted it as a spread price.

130

131

The first chart provides a sort of long term history…..

What is going on here ?!!

As one can see the skew and the spread level moved in tandem until early 2012: “up and up”, or “down and down”…Until March 2012…Until March 12th 2012…. The move as such made sense in 2012 until February 2012 to some extent actually: when credit spreads narrowed the markets were rallying, the liquidity was improving and the skew trended narrower as well. Reversely when the spreads increased, the markets went under stress and the skew widened as evidence that the liquidity was vanishing in crisis times. The end point of late May –early June 2012 is totally absurd though with regards to this quite predictable mechanism pictured above. A simple regression of the points above shows that the end point of the markets in early June 2012 was really weird (circled in red). The skew should have been trading between 15 Bps and 25 Bps, unless the liquidity would intrinsically worsen (which actually was the case) pushing then the skew even higher….. Thus if the end of 2011 was a guide the skew should have been at 30 bps or more given the market spread levels. More a skew at zero, as was the case in June 2012, had almost never been seen in the past since 2007!?! But without this absurdity, the $50-60 billion “tangible capital gain” would have been not such a thing. Instead it may very well have been a $50 billion or more liquidity reserve, ie an instant deduction from the revenues and earnings of the bank. If one remembers that the 1bp value for the somewhat $3 trillion “basis risk” is $1.5 billion, one can multiply the 30 bps by $1.5 billion and get to $45 billion in total. Here one can reconcile the $15 billion gain that Dimon mentioned in his slides of September 2010 with the ultimate $60 billion gain “net of costs” that will be secured most likely as early as June 2012 when the IG9 10yr was miraculously at zero. This is just a gross calculation here but it provides the magnitude of the stakes at play behind this “wind down” plan and the actual “skew price” for the “transfer from CIO to the IB”. One suddenly understand a straight motivation for the bank to have waited between October 2010 and June 2012 before doing this so simple, preliminary transfer of the “tranche book” of CIO towards the IB inside JpMorgan before any “off-shoring” would occur….

131

132

The last chart will show that this absurd move was really showing up through the March 12th meeting with Ashley Bacon…..The divergence occurred right around the middle of March 2012, a pattern which could only confirm that only the CIO positions were targeted. And this shoft was happening within Jp Morgan actually. Here again appears quite a compelling reason for why this quite easy “transfer from CIO to the ib” was postponed from March 2012 till June 2012; the “job” had to be finished…Indeed, had it happened in March 2012, irrespective of the lingering $300-500 million internal dispute on prices between CIO and the IB, there would have been a fixing on skew price here through the “transfer” itself. This fixing on the skew price as recognized by JpMorgan then would have re-priced the $3 trillion of “basis risk”. Then, it would have appeared that the firm was missing $50 billion in liquidity reserves since 2008. No matter what the “tangible capital” gains that was expected to happen, the books and records would have appeared to be fraudulent since 2008 actually. That was quite a scandal, one that would larger than the future “London whale” tale. The only “way out” here was for the transfer to be done at a skew price where there seemed to be “no need for a liquidity reserve”. And this is what happened but only in June 2012 after many manipulative events had occurred in the meantime…. See the chart below….

132

133

Dimon’s statements on May 10th did the job well

The 3 charts above summarize what the manipulation was meant to achieve in the markets. They show that the bank had quite a coincidental role in that. And they display what the target was at Jp Morgan itself: the IG9 skew at zero the time for the transfer and collapse of synthetic tranche positions to be finalized without having needed a liquidity reserve in the former years. If one wonders whether the bank was just profiting from this manipulation one must remember that the inflection started around early December 2011 while there existed some $200-300 million price difference about the “CIO tranche book” depending on whether one picked the CIO London prices or the JPM IB prices. There was a big issue with the IG9 10yr skew already INSIDE Jpmorgan. The tension was sky high: the IB traders closing shop on tranche trading refused to unwind with the “tranche book” of CIO. The IB traders refused to even consider quoted prices on the reason that they were NOT marked around these market prices….

The tension remained sky high next presumably so. One must remember the December 15th 2011 early and hasty year end valuation that would be followed by quite specific requests on “unwind costs” from the Federal Reserve starting on the 21st December 2011. Then the IG9 10yr started underperforming the market. The underperformance of the IG9 10yr qualitatively will broaden, irrespective of the volume traded by CIO, in a very illiquid market while the VaR of Jp Morgan exploded in proportions that had NOTHING to do with the changes undertaken in the “tranche book’ of CIO. A VaR model manipulative change occurred in late January 2012 at the very top of the firm, not CIO, so that specifically “Iksil can keep trading”. One should read the “Var History” document on that topic.

Thus everything would be done for “Iksil to keep trading” through just all the ranks of JpMorgan the firm. On February 2nd 2012, Weinstein is invited in the Jp Morgan headquarters to promote trades going knowingly against CIO longstanding exposures. Is that a well organized setup? Weinstein traded in IG9 10yr skew then as media reports would state in 2012. Between February 3rd 2012 and February 9th 2012, Macris is worried. Artajo is demoted but Drew is NOT worried. Iksil is NOT informed of Artajo’s demotion. The top of the firm again

133

134 grants temporary but indefinite limit extensions for CIO which has breached most of its limits due to the recent ordered changes of the “Tranche book” of CIO. The firmwide CFO brings his explicit support. The firmwide CRO brings his explicit support. Dimon orders to stop sending P&L reports from the IB to the OCC. Someone at JpMorgan ALSO orders John Bellando at the CIO to stop sending the CIO P&L reports at the same time. Cavanagh the CEO of the JpMorgan treasury does NOT tell the CFO of CIO that he should fill the communication gap here toward the OCC. On February 29th 2012, Weinstein apparently tries to push Ig9 10 Yr prices at the New York Close for month end. Roberts from CITIGROUP sees it and wonders as he perfectly understands what Iksil is doing on his end. Iksil advises as hard as he can to stop trading for good. Macris is concerned openly again on March 1st 2012. He calls Bacon to the rescue in a display of distrust maybe towards Drew’s tactics here. Artajo’s demotion was not enough for Macris in any event….So far the IG9 10yr skew has tightened but not much….Not enough…March 5th & 6th 2012 Drew sends a dodgy order under the mantle to Artajo. Iksil is put in the loop from his holiday spot and one wonders why. March 12th 2012 the Bacon meeting occurs: the slides were prepared without Iksil by the way. Right after that the IG9 10 yr skew artificially tightens against all historical relationships and in the meantime ALL the CIO positions start losing money, again in plain contradiction with history and commons sense. March 23rd 2012, Drew is elevating “all the way up” the issues in a pure gesture of hers as Macris and Pinto strongly suggest on their call that day. The IG9 10 yr skew tightening speeds up as Iksil warned so loudly projecting larger losses. The seminal articles will only help the skew tightening go on….But this does not go fast enough: market players would not buy into the “JpMorgan vs JpMorgan” tale for long….They need a capitulation of some sort at JpMorgan. The May 10th 2012 “moment of honesty” of Dimon in his statements brings the final hit…..The IG9 10yr skew goes to zero in a matter of 2 to 3 weeks at last! JpMorgan winds down internally through Blue Mountain after waiting a year or so…..The bank then made not the $15 billion projected in September 2010, but $60 billion instead….

The paragraph below summarizes the elements just mentioned here

a. Conclusion: Effects of Dec 15th 2011, next Feb 9th, March 12th, April 6th, May 10th 2012 i. Balance of losses during H1 2012 1. Drift of IG9 forward spread in 2012 a. Long term history of skew i. Simulated Recovery of CIO positions

4- How did the bank report the event itself within 2012 through Q2?

Thus it has been shown so far that the bank actually reported a lot of information into its public reports like the 10-Q (quarterly) and the 10-K (annually). Yet the “London whale” legend did obscure most of this information. It matters to know better how the unique $60 billion tangible capital gain “net of all collateral costs” could have escaped the attention. It has been shown so far that the bank made this genuine unique fortune right through the “London whale” events in the net “tangible equity” VS “other assets”. Thanks to the “London whale” event the “tangible capital” did not depend any longer on the value of “other assets” themselves. The gain is $50-60 billion net of “collateral losses and other legal costs” as of the end of 2014. It had been spread over 3 years….That was still a definite value creation for the firm and the shareholders as the evolution of the traded share price versus the “book value” would prove between 2011 and 213. The problem would be the “means” that were used to achieve that bonanza gain as this part will corroborate.

Analyzing the evolution of the performance of the “tranche book” of CIO, it has been shown next that the loss has been maximized through leaks to the markets that could only come from Jp Morgan itself. The March 12th

134

135

2012 meeting with Ashley Bacon was crucial especially in having the IG9 10yr skew reduced to zero. That would not be that easy to achieve however. That IG9 10yr exposure was just 15% of the risks in the “tranche book” of CIO but that also was THE key to the $50-60 billion planned gain “net of costs”. The May 10th 2012 statements of Dimon were even more important in that regard than the media campaign or the detailed market leaks of CIO’s position. These carefully crafted statements of the CEO and the bank would be the genuine catalyst of the long planned “catharsis” as they pushed most of the total loss fast into the “tranche book” of CIO.

There has thus been a lasting market manipulation here that targeted CIO. And the bank was at least participating in it (maybe passively but certainly lastingly). Whether Jp Morgan actually started this manipulation is unlikely but what is the hard evidence for that? Roberts, Goldenberg, Hubbard, Rhule, Weinstein or his sister could tell maybe. The May 10th 2012 statements and the aftermath of the March 12th 2012 meeting in terms of leaks all suggest that the bank fueled the manipulators since March 2012 at the latest. The motives for the bank are all too clear: pushing the IG9 10yr skew to zero was in the immediate best interest of JpMorgan right when it was finalizing a transfer where the “transfer prices” mattered a lot (see the Sarbanes- Oxley laws of 2003 in relation to the ENRON case about such transfers). There is some evidence of that. It may actually be found in the firm’s internal reports related to how it computes its VaR, its CRM, its IRC, its 10%CSW, its RWA overall. Some clues exist even though they remain under confidential seal in 2017.

When looking at the very history of CDS markets one will find that historically the firm could only accumulate Skew risks that it ultimately hedged with the IG9 like most of the other big dealers after 2007. For reasons that will be explained in another document, the firm could neither refuse to keep trading on this SKEW basis nor get out cleanly so. The skew was the bread and butter for all market makers in credit derivatives since 2004. This skew risk was the trigger of the last financial crisis of 2008 quite naturally. Weinstein, CITIGROUP, CSFB or Goldman Sachs may be able one day to confirm whether the rumor initially emanated from Jp Morgan top executives, not IB traders in late 2011. There was little mystery behind this rumor as such as the “synthetic tranche” market was deemed to die.

But this is another matter than the one that is developed in this document. As the bank made a fortune anyway net of the “CIO loss”, one could have expected the bank to document well why Dimon stated that the profitability of the firm remained solid even for the second quarter of 2012. Had the bank felt “clean” it had little reason to conceal the truth all the more so as Dimon had already made statements on that line of “solid profitability”. He had so compelling things to say. But he would not elaborate, right? One had rather wonder in that regard how the bank Jp Morgan reported the event in more details as of July 12th 2012 and as of August 9th 2012. This is the content of what is to come right now.

To start with there was a restatement, uttered in contrition based on a price difference that should never have lasted actually as per the valuation protocols that were in force at the bank itself since late 2007. That at least is a consensus that is corroborated by the financial history of the last 2 decades. This admitted failure is too gross to be true. This price difference certainly deserved criticism. It was way too subjective in such illiquid markets. It had occurred on the most toxic instruments of the time. It had allegedly not been monitored with the best common sense by Artajo, Drew, Macris and Pinto (see the March 23rd 2012 call and the April 17th 2012 call between Drew and Artajo)… “presumably” to borrow some words of Dimon… Puzzling!

Markets were il-liquid. This price difference therefore was justified at the very early stage of the valuation process as such if one can at last read Iksil’s communications of the time. It was well understood actually. It should have lasted for many practical reasons evoked before as far as the estimate P&L was concerned, ie at the very beginning of the daily valuation process in the firm. The cause for the price difference between CIO and the IB was simple: il-liquidity, visibility and manipulation. This price difference was certainly NOT new

135

136 witness some yet undisclosed multiple evidence of controllers’ internal reports covering the period of June 2011 till May 2012. Yes this price difference was likely already flagged by Jp Morgan controllers’ like Allistair Webster, Brett Dooley, or Shannon Warren, since September 2011.

As the global IB controller Webster, the chief accountant Dooley, and the chief of regulatory affairs Warren would suggest in late April 2012, this lasting il-liquidity for CIO called for a change in the valuation process from a standard “mark to market” process towards a “mark to model” protocol. Keith Stephan and Macris Macris attended this meeting and could confirm. Then, in any event, the firm policy dating from 2007 (see the US Senate report exhibits and search here for “November 2007”) mandated the adoption of liquidity provisions for “model risk”, “price uncertainty”, and “concentration”. The figures amounted to several $billions for the “tranche book” of CIO alone in late 2011 no wonder. Thus this “price difference” allegedly “discovered” by top managers was actually not new at all in nature and magnitudes. The very same top managers were actually aware of the issue since late 2005 at least. This is this awareness that actually had mandated the very birth of the “Tranche book” of CIO, the formulation of the NBIA of 2006 especially with regards to valuation where CIO “should” adopt the IB prices but never did in practice. This will mandate the instructions sent from the very top of the firm to adopt at CIO London an estimate P&L process that at the end of the day deliberately violated some US GAAP standards. Yet this was NOT an insane move.

These violations were not picked randomly at all. Which were those violations? CIO London would NEVER use “consensus” prices like MarkIT or Totem for one. That secured that CIO estimate P&L prices were independent from consensus ones. As such they constituted another “price source” for the bank. Secondly, CIO London would NEVER stick to a definite closing time. That secured that CIO was taking a well synchronized series of prices, contrarily to consensuses which were blind picks. As such the estimate P&L prices were better synchronized. As such that secured that this “other price source” was of a better quality with regards to criteria that were NOT favored by US GAAP standards but was favored by the bank itself.

Those 2 violations of CIO however made the daily management of margin calls and collateral postings impossible to perform on a practical basis if the IB collateral staff only relied upon CIO London estimate P&L prices. Actually they did not rely on these prices for themselves. The IB staff had to have other price sources.

The firm was fully aware of that deliberate difference back in 2006. There was a rationale behind these 2 violations: the firm wanted a well synchronized albeit subjective batch of concurrent prices to the IB ones and concurrent to the “consensus mid prices”. That CIO contribution completed the available set of prices that the firm could use for its centralized “mark to market” process. Thus CIO London provided prices that differed at times a lot from the IB prices, notoriously so as every month end reconciliation check showed it.

Thus this price difference had really nothing that was unknown to a firm like Jp Morgan. Instead the bank had wanted it, organized it on purpose since 2006 quite officially. More it was known by the regulators since 2006 and they would approve. The NBIA of 2006 was as such a requirement inherited from the Sarbanes Oxley laws, themselves strongly influenced by the scandal of ENRON (2001-2002). And NBIA of 2006 was quite ambiguous in its wording: it stated that CIO “should” use the IB prices but also specified that VCG may NOT adjust CIO prices IF CIO differed from the IB on specific positions that both units shared. The aim was precisely to secure as many “independent” price sources as possible. The stakes were huge in reputation terms. Jp Morgan had been heavily involved in the scandal of ENRON at the time. Thus through this NBIA of 2006, dedicated to this “tranche book” of CIO, the firm allowed itself to maintain price differences between the CIO and the IB that was structural, wanted as such for quite strategic purposes linked to il-liquidity in CDS markets and of course closely scrutinized by design.

136

137

One wonders than how on earth the bank could ever claim that those wanted internal prices differences had led to this contrite restatement in July 2012 for the first quarter of 2012. What a delay, 4 months, to spot what was precluded in principle since 1934 and forbidden in the practice of the “mark to market” process since 1993….

The difference of status between Iksil and his close colleagues (Grout and Artajo) provides a clue to solve the mystery surrounding this weird price difference that should not have lasted anyway. Iksil advised the very right thing about this price difference. The bank in its termination letter against Iksil would NOT dare say otherwise. The bank indeed only claimed that Iksil had “let” Artajo “order” Grout to perform a price selection that was in violation of the firm’s policies and procedures. But the bank did NOT criticize the action of Iksil on the matter as such. The bank did only criticize the last of orders of March 2012 while the structural price difference had existed since late 2006. The bank was therefore only pretending that Iksil could have prevented the change of the process that had occurred in March 2012. The bank here did NOT question the process that was in force at CIO since late 2006 and that had been decided way above Iksil’s head. That was it! Iksil could have made a miracle but only in theory here. Artajo was the boss of Grout, not Iksil. And Iksil had secured that those price differences had been elevated anyway. The bank would NOT make any mention that Iksil had nevertheless sent emails that triggered a full elevation on the matter by Drew herself on March 23rd 2012 ( see again the call with Pinto, Macris and artajo on the $250 million difference on “one position” alone). Thus in complete bad faith the bank pretended that Iksil could have countered the instruction of Drew on the valuation process change and the bank concealed the fact that Iksil had been instrumental in making Drew elevate “all the way up” the very impact of HER instruction here. She would elevate the price differences with maximum commotion on March 23rd 2012!

To be sure here, the bank did NOT blame Iksil for actually supporting this overall price difference. The bank could not do that since it was the bank’s top management that had ordered it. The bank solely blamed Iksil because he had “let” Artajo, who was both Iksil’s boss and Grout’s boss altogether in practice, “order” Grout to perform his job differently than what Iksil advised at the time. The bank would ignore the fact that Grout, Artajo, Macris acted against Iksil’s advice without letting Iksil know it. The bank would ALSO ignore that Pinto the JPM UK CEO then knew of what Artajo, Grout and Macris had created in terms of price differences. The bank would IGNORE that Pinto reneged at talking with “Bruno” on the matter as of March 23rd 2012….The bank ALSO had created in early 2012 a fake org-chart where Iksil appeared as Grout’s manager. In reality Iksil was only helping Artajo here for matters related to “holiday”, “printer cartridge”, “blackberry” changes, or other ancillary administrative tasks of no substance. Artajo remained the one in charge or “performance reviews”, “bonus”, “objectives”, “roles” about this “tranche book” of CIO. The bank knew it very well all along.

The unfairness of the termination letter is blatant: the bank changed the known role of Iksil to utter motives that were necessary to fire Iksil with maximum damage way beyond the sole economic loss. By its silence since then around what Iksil had actually advised, namely to keep reporting prices differing from “crude mids” internally, the bank actually admitted that the price difference was known structurally. Thus the contrite restatement was mostly a gesture and the bank’s behavior is dishonest in the termination letter. Regulators know it.

As Iksil would testify indeed quite consistently between 2012 and 2017 before every single authority, this price difference had been ORDERED multiple times since late 2006 and known by the very top chiefs at the bank of course in Iksil’s belief. In March 2012 Artajo and Grout did NOT operate as Iksil had advised. Their actions at times made no sense and they admitted it themselves at times also on the record. Thus Iksil simply testified to confirm those facts and their context. The context itself was also clarified. The price difference was structurally present in every daily estimate performance report of CIO London versus either “consensus” or IB prices. The authorities were thus well aware of that as soon as they would hear of what Iksil would have to say on the

137

138 matter. As of July 2013 things were crystal clear on Iksil’s testimony. This testimony would remain the same onwards.

Thus the bank justified its restatement by a price difference that it had always known of since it had ordered it already back in early 2007 and had made possible in 2006 structurally so through the NBIA. The question remains as to whether anything was missing in the bank control process to make the customary reconciliation and adjustments. Here Iksil was NOT questioned as to whether his own actions and role ensured that effectively the bank did NOT miss any bit of information. There was a very good reason for that: Iksil was NOT expected to value the “tranche book” and was therefore not involved with the subsequent controls. Grout and Aratjo were involved in the subsequent controls. All the regulators knew that too… Yes, Grout and Artajo were expected to be involved and therefore interacted with the subsequent controls. Iksil really did not know what his close colleagues had done in that regard. But Iksil would say enough to prove that the control functions could not have failed in theory. It remained to check then how they could have failed based upon Iksil’s actions or others’ actions….

Iksil could NOT prove until 2015 that, whatever his colleagues had done wrongly, the VCG, the controllers, the management had indeed understood just ALL the information needed to make the proper customary adjustments. Iksil had strongly suspected it since July 2012: all the information had been sent and therefore the restatement was a scam. That was based upon the few interactions he had had with the controllers at the time. But Iksil had no “proof” of that because he ignored what the control guys would say they had understood from Iksil himself or from others. In March 2015 Iksil could verify that they had fully understood the situation as Iksil had described it to them then. After 2015, Iksil could establish that no later than March 23rd 2012, the whole top management of the firm was aware indeed of the price differences for sure. More, the whole top management of the firm was alerted by Drew in person about the fact that for “one position” the price difference amounted to $250 million “yeah”….( see the March 23rd call with Pinto, the UK CEO of Jp Morgan and top IB chief).

Thus not only the “faulty” price difference was known but it was actually NOT moving the needle for either Macris, Drew or even Pinto beyond this theatrical elevation of Drew. The price difference was Not the issue since, as Pinto stated it, the books were NOT mismarked in March. He knew what he was talking about. Yet the CIO accusations of the “IB manipulating the markets” was “very, very, very, very” serious all the same. Thus just ALL the top of the firm backed this “$250 million…yeah” difference on one single position of CIO. This could only be the IG9 10yr position and this could only imply a 5bp difference, ie almost 2 standard “bid-offer” quote ranges. This 5bps difference in prices was going AGAINST Iksil’s advice at the time on the matter but it was backed by CIO top chiefs and implicitly by Pinto, who could not care less. Therefore the contrite restatement of Jp morgan in July 2012 was really just a fake.

Still until 2015 Iksil could NOT tell with certainty whether Artajo had or had not conveyed misleading information in early April 2012 or later. Iksil would discover the answer through the report of Allistair Webster in February 2015 (available in the US Senate report exhibits but with NO title, thus preventing any outsider to know what this report was really about). Webster’s report had been issued on May 10th 2012 quite confidentially. Here in 2015 at last Iksil would read that Webster had perfectly understood indeed how CIO differed anyway from “consensus” and the IB in its own selection of prices. Webster found a missing $307 million ultimately. He Webster the IB global controller who had just audited CIO and who ALSO was finalizing the firm-wide new valuation policy (dated May 10th 2012- effective January 1st 2012- available in the US Senate report exhibits) flagged a missing amount of $307 million….That amount would NOT cause the necessary adjustments that should have followed. There was just NO “control failure” here as of May 9th 2012 despite the allegations of JpMorgan.

138

139

Here a peculiar mismarking would therefore be deliberately engineered instead by the bank, not by anyone among the CIO “traders” of London. This website has disclosed some evidence of that already. That probably limited a lot the case of the DOJ against Grout and Artajo when these points went public on the web. Thus the restatement of July 2012 based on this price difference was a deliberate misrepresentation of the bank when one reads the controller Webster in May 2012. This is proved by the existing evidence present in the exhibits of the US senate report as said. The fact is puzzling since this was public evidence since March 2013. Was it a “technical” error of Dimon here? What are $300 million indeed in this “multi trillion balance sheet”? It should then have given way spontaneously to the revision of the liquidity reserves called “credit cost provision”. None of this occurred.

What follows will show that this is quite an unlikely ‘mistake’……This part will review what the bank placed in its quarterly 10-Q report of August 9th for Q1 and Q2 2012. This is not a comprehensive analysis, just a summary, that still will be quite long.

It matters to specify one last point. The firm on July 12th 2012 announced major changes in the way it should monitor big risks at CIO with regards to derivatives. It presented slides exposing a new set of risk limits and risk analysis that should definitely have been applied to the “tranche book” at CIO before. The bank stated that this would have prevented the catastrophe to happen. The issue here is that the bank actually took for itself all the ideas and reports that Iksil had submitted to his management since late 2011 on the matter. Iksil in January, February and March 2012 submitted again a more defined framework that would fully be adopted and audited by management (Macris and Artajo), risk control (Stephan and Kalimtgis), trading (Eric de Sangues at Artajo’s request), and be implemented by IT with Colin Edwards, no later than March 28th 2012. Why had it been adopted in the course of Q1 2012? Because this was the very framework that Iksil had based his accurate and timely alerts on. The approach was independent so far from the current valuation process of CIO and had proved quite efficient in predicting finely enough the losses to come. The resulting reports will be the one reports that Bacon and O’Rehilly will scrutinize on daily basis once they were sent to take over by Dimon in late April 2012 (since April 26th 2012). These quite useful reports were direct results of Iksil suggestions and developments.

Thus this was something that Iksil “wanted to do about it” and that was something that the bank would endorse to its full credit in July 2012. The bank of course omitted to specify the name of the author of this critical improvement in risk management. This was thus another unfair and gross misrepresentation of the role played by Iksil at the bank. As one can see the misrepresentation keeps going in August 2017….

There was not only NO new implementation by the top of the firm here since it was in extensive use BEFORE May 2012, but also these were all ideas that had been brought up by Iksil through many former and repeated reports dating back from 2011. Thus here again the bank completely misrepresented the facts and the role played by Iksil in particular on top of what it alleged in the termination letter. This last matter here will be reviewed extensively with the SEC market experts in presence of the DOJ in 2013. So if one compounds already the misrepresentations around the “price difference”, the absence of adjustment or liquidity reserve, and the misrepresentations the risk analysis framework one has a clear idea of what is going to be said in what follows…..The only accurate thing in the bank statements on the 3 matters was that the “tranche book” at CIO had just NO limit as such. This is weird when one realizes that since 2007 this “tranche book” standalone VaR weighed more or less 40% of all the VaR of all Jp Morgan….The post implementation review of the NBIA of 2006 had never been started…Who cared about that absence really other than regulators?

139

140

a. The actual $6 billion loss attribution on CDS trading at the whole firm (different from CIO)

This part is certainly the one that everyone is focused about: “was it really a $6 billion derivative trading loss at the firm?” It was explained that anyway the reader of the 10-Q or 10-K reports would only see what the bank “saw”, not what the counterparties saw on their side in price terms. Despite all the arguments conveyed above, one may believe still that genuinely some executives at the operating committee or at the ALCO stage “saw” still a ballooning loss without understanding it. One has a starting point that the bank itself gave through its own restatement. It was a shame for JpMorgan, more than an “embarrassment” for sure to restate earnings. That event alone could induce tons of class actions. How much was this ‘shameful trading loss’ worth alone?

There was a “trading scandal” reported regularly through the media no doubt. The issue had been flagged as originating in this “tranche book” at CIO. It was caused by a “French trader off-the-chart wearing black jeans and called the Caveman at times”. People had obtained this critical information thanks to the diligent enquiry of Greg Zuckerman at the WSJ in particular. The “London Whale” may have been just another casino “whale” right? Sum up the clichés and you have the official storyboard of the $6 billion loss. Dimon would still call it again ‘a mistake that companies do at times’ in August 2017 in an exclusive CNBC interview. The scenario does not fit in many places though already, witness the “price difference that should never have existed”, the missing $307 million adjustment or the “risk framework” that already existed (as mentioned above). It is a weird scenario.

For example the firm had had a quite stringent valuation process for years if not decades in 2012 (remember the SEC annual report of 1992, the Mark to market process and subsequent events). One could see that the “gross notional” for the “tranche book” of CIO were netted according to the “market consensus price” since 99% of the “tranche book” trades were gathered in “legally enforceable” netting agreements. Any price difference between CIO and the IB was reconciled at an early stage through this netting process of “legally enforceable” contracts. Granted the bank is totally opaque on the “’basis risks” but this is NOT an issue as far the “tranche book” positions were aggregated firm-wide with other positions having the very same contracts at play at the IB or elsewhere in the bank. It was also shown in the early part of this document that the second quarter, after restatement, was ending with a very benign loss of some $385 million for credit derivatives being under “legally enforceable” netting agreement, not a $6 billion figure. This was the case for just 99% or more of the trades held in the “tranche book” of CIO. Thus CIO alone reported a $4-5-6 billion loss truly so on those exposures through the second quarter of 2012. This loss at CIO was genuinely caused by observable market price changes. The bank had “struggled” as the reports of Zuckerman and others would say….There must have been high trading costs, right? But the bank itself only recorded a “business as usual” $385 million, as far as credit derivatives and “consensus prices” were concerned for the firm as a whole through a “legally enforceable” netting, ie the context of the “tranche book” of CIO. The shadows hanging over the embedded “basis risk” should not apply as far as the “tranche book” loss was concerned. Thus this $6 billion straight loss coming from market price changes directly impacted the gross receivables and gross payable amounts on credit derivatives. It should not have ended being just $385 million IF the firm really had lost money here.

This $6 billion “CDS trading loss “ showing in “principal transaction” is therefore a pure “input” from either the risk systems of the firm (not excluded yet) or from the senior management of Jp Morgan alone. Which one did the “job” here of conveying this “$6 billion trading loss for the firm” stamp overriding the market realities?

It must be reminded that the ALCO did not materially change the result for Q2 2012 on “credit derivatives”. From the $385 million loss, the ALCO would make it slightly lower at $360 million actually. Aside from that, the firm did NOT make any mention of a change of its risk systems that would justify this $6 billion

140

141 rubberstamp “trading loss”. One is left with senior management at the ALCO stage. Why did the senior management then placated a $6 billion trading loss like this in “principal transaction”? What would be the rationale since neither market prices nor in-house risk models were the trigger alone? Let’s assume for a moment that this “tranche” book was a hedge that had been mismanaged indeed by someone at the firm. Let’s assume that, against the statements of the bank, the statements of the medias, the statements of all the regulators, this “mismanagement” did NOT materialize actually in harmful price changes for the firm on its trading performance. Where did it materialize then if it ever did? Then at the ultimate ALCO stage, top bank chiefs would have been witness of a loss of a special nature indeed after all the prices had been “corrected”, after all the collateral management operations had been performed, after all the “fair value” options had been performed, after all the performance attributions had been processed through the firm’s in-house mathematical risk model…They may have kept the real source of loss for confidential reason pertaining to the business. CIO and “traders” would then have just been scapegoats… This would have been quite unfair but not too dishonest on the part of the bank so much.

This is then, at the very end of the valuation process, that the ALCO may have “discovered” such a nasty almost invisible gross mismanagement…This is in essence the motives that supported the termination letter that the bank sent to Iksil, Grout, Artajo, Macris and others maybe. That was a ‘collective mistake’ and the “buck stops at me” Dimon was reported saying then…. And a $6 billion figure is really no big deal after all for JpMorgan. It could have been just $4 billion or even $8 billion likewise. Given the sheer size of Jp Morgan one may “guess” that the ALCO and the operating committee members had to deal with $billions of adjustment every quarter anyway. It remains that the ALCO would have in good faith reported a $6 billion loss for the firm, not on “credit derivatives” but a failing hedging strategy that had been done at CIO on credit derivatives. The ‘flaw’ then was that this hedge had been “mismanaged” and lost against assets that had made no gain. The ALCO customarily had to shift $ billions from one quarter to the next. And here they had a $6 billion popping up like a rabbit out of the hat to their genuine dismay….It is time to review this scenario seriously, thus forgetting the march 12th meeting, the manipulation evidence already mentioned and the May 10th statements….

The account around the SFAS 107 ultimate adjustment provided some color on the matter: in 2008, the ALCO had played with $15.5 billion of inflated profits…In 2009 the ALCO again had injected about $7 bln of intangible profits as well…. From Q4 2011 to Q1 2012 the ALCO had actually injected $5 billion that were way sufficient to allow the firm to report a $5 billion earning profit for Q1 2012. The ALCO could thus inject the current quarterly profit or even more at the last minute of the valuation stage. Such was the order of magnitude of subjectivity for the big US bank JPMorgan at the ALCO stage. And the watchdogs were watching the SFAS 107 rule, weren’t they? The accounts for Q1 2012, printing $5 billion earnings profit by means of a $5 billion cumulated adjustment from the ultimate ALCO stage (from Q1 2011 to Q2 2012) was a “tempest in a teapot”….. The ALCO had better make the minimal adjustment whenever possible as good common sense suggests. And the sooner the better…So the ALCO certainly did its very best. But the firm would obviously NOT disclose how the ALCO processed over time. It would not either for Q2 2012.

Still, based on the section about “fair value measurement” that is given in the annual reports and on the “fair value election” tables, one can proxy how the ALCO adjustment evolved over time. A lot of information is available and allows to cover the whole valuation process. Some parts are missing. What is certain is that one may be able to observe how subjective the ALCO could be before the stage of the SFAS 107 rule. Thus the reconciliation that comes is not complete and conveys obvious remaining uncertainties. Yet, the uncertainty itself can easily be flagged and quantified. Given the disclosure rules one should be able to highlight very clearly the $6 billion negative adjustment that the ALCO alone set on the “principal transaction” to sort of “document” the clamored “$6 billion trading loss for the firm”. The issue is that this $6 billion ALCO-led “trading loss” is in plain contradiction with hard figures so far. History again is quite a good guide on that

141

142 matter. To be sure the actual “trading loss” on credit derivatives for the whole firm on the very products and contracts that the “tranche book” of CIO used was $385 million, NOT $6 billion.

The way to spot the “actual” loss in “trading” that the bank executives actually “saw” along their own valuation process then is definitely to look at what they themselves flagged, despite the unchanged earnings at the time of July 2012: the “principal transaction” line. Granted the firm is “complex”, the fair value measurement” process is “complex”, the models are “complex”, the firm is self-admittedly “opaque” and so on….But, as explained, the firm produces figures and makes ultimate adjustments at the ALCO stage that one can infer.

It matters to rebuild here the way the firm actually reports its own figures. Among the “additional notes to consolidated statements” the firm provides figures as to how the fair value election stage influences other business units in the banking group. With good common sense one can infer well what likely has been transferred from the gross performance of credit derivatives into other parts of the firm. If one wants to secure that the associations are right, it suffices to reconstruct what the process must be, what the totals would be and how they fit together over time. It cannot be perfect but common sense helps a lot.

Thus the bank starts from the beginning, ie different price sources (IB, CIO, counterparties, ICE, TOTEM etc…). Next the bank nets the contracts of the exact same kind through legally enforceable agreements. There might be microscopic price differences here IF the netting process does NOT shrink the notional amounts materially. The CIO and the IB most often, like 80% to 90%, have “open trade” on the instruments present in the “tranche book”. Here with certainty there can be only one price inside JpMorgan. Next a first set of reserves called “initial collateral requirements” is made. It should not apply to the trades between CIO and the IB as both units are inside JpMorgan: there can only be one price here. Then the bank runs a risk analysis and performance attribution based upon “identical underlying” risks in order to match “cash exposures” along with “derivatives exposures”. Then the bank through risk management teams AND finance department devise the strategy and the reserves. The in-house mathematical models have not been changed apart from the VaR one….

Maybe that was here, through the “fair value” election process whereby risk models mingle with routine reserves that the $6 billion “trading loss” actually appeared…. That should be the case if the scenario suggested above had applied in July-August 2012 in the eyes of the ALCO. The ALCO starts being involved indeed at this stage as the firm reports. Next, post the fair value election process, the ALCO approves the reserves, analyze the capital provisions and defines the future strategy of the firm. At this stage the ALCO sets its adjustment both on credit derivatives, on reserves of all kinds, on capital provisions, and adjust the assets to the liabilities in a way that will affect the reported earnings of the time. Many stages here are clearly confidential, especially the ALCO inputs. One can recognize that AFTER the mechanical price collection, AFTER netting, AFTER risk attributions through the “fair value election” process, there is a shadowy stage all driven by ALCO. One may expect the ALCO to create itself huge swings year on year on the adjustment that the ALCO alone sets for the credit derivatives or for the “principal transaction” figures. This would be fairly justified given the magnitudes that were mentioned in relation to the SFAS 107 rule. What follows will show that this is not the case so much apart from very few exceptions.

The table below shows that many figures are produced indeed about this “fair value election” process…..One has enough data to proxy rather well how the ALCO stage “normally” is run here. This table displays series of figures and squared lines that intuitively explain how the transfer of performance from the “the gross netted trading performance” of credit derivatives into other business units, led to the ultimate “principal transaction” that reported the “trading performance” as disclosed by ALCO in the very final hand. IN other words, the ultimate ”principal transaction” figure reported the “trading performance” as the ALCO wanted it AFTER some

142

143 routine “attribution” had been added by firm-wide risk models as they had been performed based on the “cross business Fair value election process”.

For the sake of clarity this table is just a sample of the data collection that is designed to show in plain transparence how the coming computations will have been performed. This will help those who would like to check the coming other tables. As one will see the selection of items that likely impacted the credit derivatives ultimate “principal transactions” value is quite intuitive since the firm does not explain much itself. Thus this table is here for the sake of transparence as far as this report is concerned…It matters because, given the many information gaps, one can only spot the $6 bln impact of ALCO in what is usually a pretty steady adjustment at this ALCO stage for credit derivatives. At the top of this table below one will see the “total Fair value impact” that likely came on top of the purely market driven performance based on price and “legally enforceable” collateral management. This impact here can be understood as the output generated by mathematical internal risk models coupled with more strategic “options” that the ALCO already took in terms of “fair value” reporting.

As one will see, these individual figures varied quite a lot form one year to the next at times for each bucket, emphasizing the notorious “complexity” and the highly interconnected role played by credit derivatives in the bank. This alone shows that the “tranche” book of CIO, weighing about 40% of the VaR, could not be left as the bank pretended it was, ie “poorly monitored”, “poorly executed”, “flawed”, “poorly reviewed”. One should notice also the peculiar spike in Q3 and Q4 2011 of this “impact” that lines up with the $30 billion jump in the bank cash drawers contemporaneously so….This table below is here just to display the Fair value election impact that comes into the equation before the ALCO is determining the ultimate “principal transaction” figure. One will notice the many spots that are involved potentially with the activity on credit derivatives. This is not a mechanical connection but one that likely existed once the “tranche book” of CIO had been designed to hedge the bank against systemic risks for the balance sheet.

143

144

This quite long table is just here for the sake of transparency and just to indicate the choices that were made to produce the next tables. The following table will show indeed that, as a reconstruction based on those choices, one can most often replicate pretty well in fact the evolution of the ultimate “principal transaction” through this intuitive use of the “faire value election” figures. At least one can do that most of the times unless the ALCO

144

145

“sees” something of a peculiar nature. And therefore one can see when the ALCO specifically injected its dose of subjectivity that did NOT pertain to market prices be that on derivatives, one hedges or on other assets….

Indeed one will see that when adding the “gross legally enforceable netted amount” to the selected “fair value election” figures displayed above, one gets to a figure that is close to the “principal transaction” ultimate figure. It is not quite the same all the time but the part missing, ie the ultimate ALDO adjustment set post the “fair value election” process, is often small and reported below as “Principal transaction chge done by ALCO”.

One can see it on the line labeled “Principal transaction chge done by ALCO”. Take Q1 2012, or take Q1 2013, Q2 2013, Q3 2013, Q4 2013, Q1 2014 and so on, the “assumed” ALCO final adjustment is minor really. However the ALCO quite uniquely decided to impact negatively the “principal transaction” performance of credit derivatives starting in Q2 2012 until 2012 year end by some $6 billion. This $6 billion ALCO adjustment here is the only “mathematical reason” therefore that is supporting the tale that indeed this “London whale” was a $6 billion trading loss outright for the firm as shown in the “principal transaction” total. The coming table here thus shows that, when one looks at the reported principal transactions figures in credit derivatives, the year 2012 was peculiar. It is clear how the ALCO alone did impact this figure post the “fair value election” process. The $6 billion loss is a pure writing of senior management, not a market reality for JpMorgan as a firm even through its mechanical risk analysis and performance attribution across businesses:

The numbers that are circled above in red are computed as being the reported “CDS principal transaction” minus “Total Fair Value Impact” (see former table) and minus “gross CDS Fair Value change Year-to-date”. This suggests even further that this $6 billion is just a “$6bln stamp” that was placated on the routine figures for the occasion only. Thus if one sort of re-writes the implied equation here one has:

“reported CDS principal transaction” = “gross CDS Fair Value Year to date” + “Total Fair Value impact” (from fair value election process) + “principal transaction chge done by ALCO”.

That last one figure drove completely the reported numbers and was NOT representative of the reality conveyed by the bank’s net positions as they had been impacted by “consensus prices” or “fair value election”. Thus the ALCO members surely DID not reflect the actual net effect on the bank JpMorgan of the $5 billion loss that had occurred at CIO for sure and as the bank itself “observed it” in the markets through the lens of its counterparties. If only because they were officially “concerned” all the watchdogs must have watched that out, that JpMorgan was “only” losing $6 billion on paper which allowed the bank still to keep its earnings as originally announced for Q2 2012…..

This equation above, however approximate, replicates the way the firm describes its ‘fair value measurement’ process. Of course some intermediate steps and adjustments are missing. But the table above shows that this ALCO-led adjustment shall be quite benign normally as per this modeled reconstruction, ie other than for Q2- Q3-Q4 2012 actually. The “assumed” figures from the ALCO were and would be in the range of $300 to 500 million out of these 3 quarters of 2012. This is a strange coincidence that this $6 billion adjustment fit right for the reported “principal transaction 10-Q” number that by an extraordinary coincidence led people believe that

145

146 anyway the firm saw a $6 billion trading loss in Credit derivatives then. But one may still argue that in the past years the ALCO had done other larger adjustments. The following table will show the longest possible record on avail since the details of the fair value election was showed since 2010 only:

As one can see actually, it is only in the first year of reporting that the ALCO had set a growing positive adjustment of $6 billion by another weird coincidence in 2010… The earnings looked good in late 2010, almost too good to be true if the regulators reaction was a guide. They were officially showing concerns around CIO and its “SCP”. The bank printed high profits then but there must have been questions around the “principal transaction” adjustment of ALCO for credit derivatives right? It was NOT a negative number but quite a high and steady positive number then, all along 2010. This is when regulators started expressing quite deep concerns by coincidence again…. Still, this adjustment then went growing regularly in the second half of 2010 and this came along with a heightened scrutiny from just all the watchdogs. With no surprise, one sees next that in 2011 the adjustment looks benign….until Q2 2012 once again….The ultimate Fair Value change impact by ALCO ( as assumed here) is also provided for the sake of transparency although, the fair value change alone ultimately does NOT translate the own perception of the firm about its trading performance. The pattern on the table above is all the more obvious with regards to the deliberate but quite subjective decision of the ALCO to state a $6 billion “trading loss” on credit derivatives as there was no such thing in reality. Here regulators would just never blame the ALCO for this quite subjective $6 billion loss that senior management created for the occasion. The ALCO was conservative wasn’t it? Were the regulators simply supporting such descriptions? Why was that? Volcker rule?

The next chart is meant to plot the history of the ALCO adjustment on CDS “principal transaction” that shows plainly that there was a deliberate will to draw the picture of a $6 billion trading loss in 2012 that was just that a “picture”. That “will” was discussed with the watchdogs then and has been endorsed by the watchdogs until now:

146

147

Regulators warn and the ALCO goes to zero

The ALCO manufactures a $6 bln trading loss. Regulators back this…..

b. RFS standing for “Retail Financial Services”

Thus while the bank in July 2012 manufactured a role that Iksil never had, caused here against its employee maximum damage going way beyond a financial penalty, manufactured a restatement based on a price difference that should not have lasted longer than a day, manufactured a “risk control failure” that never existed in 2012, and manufactured a “$6 billion trading loss” on CDS out of the blue at the ultimate ALCO stage in quite a subjective manner, the bank ALSO provided its own form of explanation still. How reliable was it?

Indeed, all this was “surprising” to say the least. The bank produced earnings just as expected (ie right at the level where they were announced before the scandal erupted) then despite this sudden, allegedly unwanted $6 billion trading loss “because of CIO London traders”.

It matters to see now how accurate this description was, after all this contrition about the CIO “massive and flawed” trades. In that this part addresses specifically what Carl Levin himself raised on his website when he commented in September 2013 on the “settlement” that the bank had reached with the US authorities then : “Car Levin 19th September 2013 : « the whole issue of misinforming investors and the public is conspicuously absent from the SEC findings and settlement » “The size of the penalties is testimony to the great damage risky derivatives bets can do, and that's important. However, the whole issue of misinforming investors and the public is conspicuously absent from the SEC findings and settlement. Our PSI investigation showed that senior bank executives made a series of inaccurate statements that misinformed investors and the public as the London Whale disaster unfolded. Other civil and criminal proceedings apart from this settlement are continuing, so there is still time to determine any accountability on that matter.”

147

148

What about the misrepresentations that had been effectively conveyed to investors and markets in July 2012? Regulators had failed no doubt and would fail even more later on as was pointed out already. And yet they were watching this out closely for sure after June 2012. How much were these “explanations” misleading or mischaracterizing the facts? ….

A further analysis on the slide 1 of financial results that were produced between July 12th 2012 and August 9th 2012 shows what the bank stated so voluntarily then. What did the firm announce on July 12th 2012, placing all the blame upon ‘CIO traders’ and their marks? The table below shows the official message of JPM for Q2 2012:

The figures above are the ones disclosed for the second quarter of 2012 alone. One figure stands out at first sight. This is the $2.1 billion gain that soon the curious reader will see comes from “RFS”. The maths above are overly simple for anyone to carve out the “picture” in anyone’s mind. Figures balance one each other as a whole: the incident indeed was just a “tempest in a teapot”.

To retrieve the total $5.8 billion headline year to date loss, one has to remember that in Q1 2012 the bank made a restatement due to “price differences”, worth of $660 million, that led to a loss of about $1.4 billion for the first quarter of 2012 at CIO. Here the bank shows how the $4.4 billion Q2 alleged trading loss at CIO was finely balanced in fact and how therefore a genuine ‘miracle” occurred at Jp Morgan “in hindsight”. One may notice that the bank confirms that the performance of the “tranche book” of CIO was “seen” as balancing the result of “other assets” through –for example- the reference of the “bear stearns-related first-loss note”. One will notice that the performance of the “tranche book” of CIO was also related to the “basis risk” through the reference to “DVA gains” and the reference to RFS via “mostly mortgage and credit cards”. One will also notice that the performance of the “tranche book” of CIO was also connected to the “strategic liquidity reserve” of JpMOrgan that CIO had the mandate to invest in the markets. This is what is reliable in this table above. The rest is NOT.

The $5bln projected quarterly profits for the bank remained finally a $5bln quarterly profit DESPITE the quite sudden and unwanted $5 billion allegedly unpredictable ‘trading loss of CIO’.…. While people had been prepared to expect a “$5billion expected routine profit MINUS a $5billion sudden prop trading loss= $0 billion profits” outcome, there was simply a “$5 billion profits finally as initially announced- don’t think”. Just that situation here in July-August 2012 looks too miraculous to be true. The bank again made no real effort to explain how this magic balance was achieved over time, far from it. Only the confusion would be “documented”.

Was this book a hedge or not at the end of day? The bank would just NEVER provide supporting documents for this hedging function. Was it because this “tranche book” was NOT a hedge? Looking at the table above, it very much feels like the answer was “No this SCP was NOT a hedge- We just at Jp Morgan got lucky to get out unharmed – We here present you with a nice fairy tale- read my lips…”. And therefore it also strongly feels like

148

149 the firm scratched across all its provisions at the last minute to save the face here….This picture is totally misleading as it will be shown now.

How were those $4.4 billion quite real “trading losses at CIO alone” so coincidentally and timely balanced within Jp Morgan as per the top executives of the bank?

CIO realized gains on securities to recover $1 billion allegedly…Really? Well CIO had actually a $10 billion unreported gain in its $360 billion worth of investments. CIO had realized some gains then because the securities in question had reached their maximum projected gains ahead of time and could all be sold back at good prices, full stop. If really the CIO securities gains were to offset the “’tranche book” losses, enough was to re-price the assets constituting the $10 billion unreported gains then. But this would have triggered very embarrassing questions as to the price uncertainty surrounding the $360 billion worth of investments managed by Drew at CIO for Dimon. CIO investments were supposed to be really liquid as they were deploying the “strategic liquidity reserve” of the bank. They were deemed similar to “cash at hands” at any point in time. Were they “liquid” really? That was not quite the case in reality if one realizes that these $10 billion unreported gains at CIO were mostly due to price uncertainty. $10 billion out of $360 billion (more than 3% of the total) suggests that the value of the “strategic liquidity reserve” was uncertain actually to the tune of $10 billion. So this $1 billion securities gain was NOT here to balance the loss in the “tranche book” anyway….That was what CIO could actually realize over the $10 billion unreported gains measured on the $360 billion worth of CIO investments. Regulators knew it once again. Other gains came coincidentally to complete this dubious one….

The main windfall “gain” on the “explanation” table above came from a “benefit from reduced loss reserves, mostly mortgages and credit card”…..One can quickly infer that it comes from “RFS” in fact. Another windfall gain at the Investment Bank on “DVA” and another profit from a Bear Stearns “related” operation dating from 2008 providentially came right at this time in Q2 2012. They all came right then to finish balancing this allegedly unwanted catastrophe. None of those counter-balancing gains appear to be related to the CIO “tranche book” that was supposed to be a hedge for the firm. What a convoluted way therefore to guide the analysts into thinking that indeed that “tranche book” was just a hidden “prop trading shop” necessarily run by the “French trader wearing black jeans”.

Now the long silence and entertained ambiguity of the PR staff of JpMorgan since April 5th 2012 finds a clear explanation: they did not want to admit it but finally said so implicitly. This ambiguity here does not look at all unwanted but instead looks quite strategic for a firm who wants to avoid class-actions. “This is a prop trading scandal but we at JpMorgan are fearful to confess it”. That is all clear between the lines in the table above, right? This serves so well the cause for the Volcker Rule and the cause of the bank altogether…..

Then common sense should have brought up some questions here anyway. One must figure then for example that, absent the “London whale” sudden scandal, the earnings for Jp Morgan would NOT have been $5 billion as projected but $9 billion instead or even more for that second quarter of 2012! This is just common sense and simple maths. Another point should have been made that regulators were scrutinizing this “explanation” above and MUST have checked all this quite thoroughly as a $9 billion quarterly profit figure “out of incident” for Jp Morgan at the time was a miracle indeed. And, mind you, the regulators approved all those public statements of Jp Morgan. They never sued the bank executives here on a personal basis As Carl Levin pointed out in September 2013….

Was this truly such a miraculous alignment of the planets being favorable to Jp Morgan through that very same quarter when Dimon was allegedly plagued lethally by this “London whale” scandal? No, it was not such a miracle at all and regulators saw it plainly, whatever the Volcker rule outcome would be. It is enough to

149

150 remember the $30 billion cash pile that had popped up in the 3rd quarter of 2011. That was NOT a gain. That was NOT a reserve. That was NOT a divestment. And it is enough to see that a $30 billion cash pile stood around still in Q2 2012 while the “tangible equity” got way bigger than “other assets” by a $20-25 billion amount on top of that…. At last…

All is just uniquely beautiful irrespective of whatever happens at RFS actually. But RFS, “Retail Finance Services”, conveniently plays the role of a red carpet in the Volcker Rule arena of the time. Indeed “Retail Finance Services” is supposed to deal with the man on the street, thousand miles away from the arrogant “French traders” or Wall street shenanigans…. The case of RFS is quite illustrative of that new “smokes and mirrors” game involving now directly all the regulators when one digs further in the 10-Q reports of the time.

As hinted before through the “fair value election” process description and through the “September 2010 slides of Dimon”, it turns out that RFS was among the business units in the firm that was very involved with the “wind down” plan when it was about to re-direct the gross derivatives performance. A large part of RFS final result came from trading in derivatives through the “fair value election” option as said. More, the mortgage future servicing fees were actually hedged by the CIO of Ina Drew in New York since 2005 at least. In 2006, the specialist named Brian Egnatz would be the first active “partner” of Macris in designing the shape of the future “tranche book” of CIO to start hedging the “strategic credit risk” of the bank’s many loans. The connection was straight, as far from obscure prop trading as one could imagine, provided the bank had explained it. Thus the relation was structural between RFS and the “tranche book” of CIO conceptually and historically: the mortgage interest and credit “residual” risks were all monitored and managed through the CIO since the very early days of CIO in 2005.

But why did JpMorgan just NEVER make this clarification to a broad audience eager for documentation of the “strategic hedge”? Was it so difficult to be convincing when tens of emails or calls between Egnatz, Drew and Macris existed in the Bank emailing database on the matter?

Now the stated $2.1 billion gain in Q2 2012 looks to come from mere “provision releases”. What a pity and what a diversion from reality! As Dimon would state before Congress in June 2012, this “tranche book” had “morphed” into something that went out of control… As described before the book had “morphed” actually into what “Jamie” and “Ina” had wanted to have months and quarters before the “London Whale” scandal would emerge. Here in July-August 2012, this new “morphing in hindsight” from a long-standing and strategic hedging pattern in the markets into an opportunistic “reserve release” statement feels like cooking the books actually. It is not quite that. It is just misleading. Digging into the actual 10-Q report, one would find out that there was no such net ‘reserve release’ overall but instead small provisions that were released only at RFS through Q2 2012 that were more than compensated elsewhere.

Thus, other than this misleading characterization in the table above, the actual figures on provisions in the 10-Q report convey a different story. As a deeper study of RFS shows in particular, focusing on ‘Mortgage fees and Related income’, credit cost provisions had increased albeit less than in the past quarters. Net there was a net release at RFS on past provisions. Yes some credit cost provisions were kept to the tune of $500 million, lower than in 2010 and 2011 but being here in line with an ongoing reduction trend for existing reserves that had started in 2011. Once again the connection between RFS and the “tranche book” of CIO was blurred. That is where one may consider a “release” of some sort of about $500 million but unrelated actually to the “tranche book” anyway. How confusing this “explanation” was!

So there was a net “500 million release” in credit costs at RFS. Granted….However some litigation reserves and tax provisions were added that amounted in all to $2 billion. Thus the facts are that the firm increased its

150

151 provisions on a net basis by $1.5 billion where it states that it ”had released” reserves to the tune of more than $2 billion. Why then put “RFS” in the loop when “explaining” the ultimate miracle around the “London whale trading loss”? All this “gain at RFS” was actually supported by an unprecedented and quite material increase in the ‘net production revenue’ that came actually from ‘related income’ extracted away from the typical loan origination and routine fee collection. It matters now to identify as best as possible the actual source of this massive gain at RFS that indeed was directly related to the fate of the “tranche book” of CIO. The following tables are going to display all this as it was reported in the 10-Q for Q2 2012 actually….

The reported $2.1 gain of RFS, was NOT an isolated one-off quite coincidental gain here associated with “credit costs” reserves. RFS will record a total $10 bln gain in the following 12 months. But when analyzing the very descriptions of the bank Jp MMorgan, it is hard to justify it by the standard operations of “RFS” alone. Was it an ongoing release of reserves? No, it was NOT. What was then the source of this overall $10 billion gain? The analysis must start from the highlight made by Dimon that RFS released $500 million on ‘credit costs’ (not 2Bln as suggested) paradoxically. The fact is that the firm took other provisions that overwhelmed this quite apparent release (litigation $400 million and tax related expenses $1.5 Bln). This release as such will be described a bit more below. The other fact is that RFS ‘pre-provision revenue’ exploded higher. This likely allowed JpMorgan to actually take all those additional provisions that were taken so quietly elsewhere in the bank away from investors’ sight.

The table below shows how this performance of “RFS” breaks down. The firm highlights ‘credit reserve releases’ as well as ‘mortgage fees and related income’. The sea change at RFS is on “Mortgage Fees and Related income”. This increase in revenue is NOT due to “mortgage fees” as such. One is therefore left with “related income”. This “related income” was managed through the IB for basic hedging purposes and through the CIO next for “global risks management” of the future premiums, ie the future revenues. The ‘credit cost’ release looks small compared to the net reserves of $1.5bln that were actually additionally taken in the course of Q2 2012. More one can notice in red that the ‘income before tax expense’ increased a lot in Q1 and Q2 2012 as a result of the sea change in “related income”. This most likely occurred in the walls of CIO and of the IB…. So it is worth both verifying where this comes from and check whether those ‘credit cost’ releases of $555 million were overdone or actually minimized….

The ‘credit cost’ provision release is the easy part. It looks to be the booster of the net income before tax expense as circled right above. This “credit cost” provision release requires a longer term history to show that this release was long expected in fact, ie unrelated to the spotty scandal of 2012. This had nothing to do with the “London whale” event as such. Thus in any case this mention to “release” of reserves for RFS business was misleading in the “explanation” table.

The release was long planned and long expected. The firm acquisition of Bear Stearns and Washington Mutual had brought in many non-performing loans indeed that had mandated massive provisioning in 2008 and 2009. The loans came to expiry as the firm explains and the reserves turned out to be in excess in 2012 predictably so. Thus this release had very little direct connection to the CIO hedging book current travails and was by no means overdone. Instead it looked quite timely and expected. As such this “provision release” from RFS should NEVER have been brought up as a sort of “last minute” balance against the CIO loss since it was already in the cards at the end of Q1 2012 when the Q2 2012 first earnings projections were made. Once again, contrary to the

151

152 suggestion of the slide of July 12th 2012, there was no such need to inflate reserve release on the credit bucket at JPM to balance the sudden CIO loss of Q2 2012.

The chart bellow shows the quite predictable and long term trend for “provision for credit costs”:

Provision for credit costs

5000

Provision for credit costs 4000

3000

2000

1000

0 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 2008 2008 2009 2009 2009 2009 2010 2010 2010 2010 2011 2011 2011 2011 2012 2012 -1000

As to the sudden increase of ‘mortgage fees and related income’ at RFS in 2012, the chart below shows a compelling story:

Mortgage fees and related income 3000

2500 Mortgage fees and related income 2000

1500

1000

500

-500

Q3 2010 Q3 Q3 2008 Q3 2008 Q4 2009 Q1 2009 Q2 2009 Q3 2009 Q4 2010 Q2 2010 Q4 2011 Q1 2011 Q2 2011 Q3 2011 Q4 2012 Q1 2012 Q2 2012 Q4 2013 Q1 2013 Q2 2013 Q3 2013 Q4 2014 Q1 2014 Q2

Q1 Q1 2010 Q3 2012

-1000

One should look at the bump of revenue that started in Q1 2012 and ended in Q2 2013. That is quite coincidental with the dismantling of the “’tranche book” and of CIO altogether right then. As said CIO had been

152

153 hedging the interest rate global liquidity risk and the credit global liquidity risk associated with the “servicing rights” that RFS collected routinely. “MSR” stands for “Mortgage Servicing Rights” and was the starting point of Drew, Macris and Egnatz back in 2006 to shape the future “tranche book” of CIO. It is thus quite natural to speculate that some embedded unreported gains of CIO went to the credit of RFS while CIO was being dismantled. This is just a scenario but it would have supported well a better “explanation” for the “miracle”. Truly this scenario would NOT have helped a lot the authorities in their push in favor of the Volcker rule. Truly also that scandal could NOT had been attributed to “traders”. It would have likely been attributed to “regulators” actually at the end of the days.

It is worth looking twice at the chart above. Apart from the unique spike of Q1 2009 and the trough of Q1 2011, RFS produced per quarter a net profit of $600-700 million. But starting in Q3 2011, RFS started making around $1 to $1.5 Bln net gains per quarter until Q4 2013. This additional gain is compounding to a total estimated extra profit of $10 bln overall through to the end of 2013. Where did this extra gain come from? Was it a transfer of the $10 billion of unreported profits that were actually sitting in the price uncertainty of the $360 billion worth of CIO investments? One struggles to find an independent explanation for this surge of profitability of RFS in the context of the time.

Historically indeed RFS and the business of servicing loans were more profitable as rates went up (not down to zero as was the case since 2009 actually). Thus the boost in RFS revenue could NOT be explained by the macro economy or the level of rates. In fact those factors suggested a lower revenue for RFS than the historical average, not a higher one. Banks regularly complain indeed about the fact that due to QE and zero rate policies, their lending margins are all reduced and the servicing fees are squeezed even more…. It is ALSO quite intuitive that at zero rates the lending margins tend to squeeze down to zero soon or later. This is the very vow that central bankers repeat to justify their own “zero rates” policy. It has to “work” somehow, right?

How could RFS therefore achieve such a 100% increase in its profitability here then? As pointed out before, this was all done through “related revenues” ie “risk management” trading revenues. Starting in late 2011 RFS was making almost $1 billion more every quarter while, as the former chart shows, the brick and mortar production revenue was more in the range of $600-700 million and capped here. How could it more than double up by means of “risk management trading techniques” on derivatives?

The next table displays the YoY moving average to show better the swing that occurred across the end of 2011. There is here a peculiar timely coincidence with the closing of the tranche business of “Credit Hybrids” and the long “wind down” plan of Dimon where the “tranche book” of CIO was to be “off-shored” and “dismantled” (or “taken down” for the OCC and CFO alike). All this plan would start being finalized after a quite simple preliminary transfer of the “tranche book” of CIO to the IB.

The moving average displays a definite sea change albeit short lived. It is not a stretch of the imagination if one assumes that Dimon, Drew and Zubrow and Hogan could predict that temporary sea change as early as 2010 about RFS as the “basis risks” would be cleared across the firm. RFS must have been a big source of “basis risk” since it was NOT a trading unit but still was a big “client” inside JpMorgan for derivatives trading. It was so big in fact that it was the main “client” of CIO new York in 2005 and so big that its “servicing fees” had been the starting point of Macris, Drew and Egnatz to begin designing the future “tranche book” at CIO.

153

154

3000

2500

2000

1500 Mortgage fees and related income 1000 Yoy Moving average 500

Q3 Q3 2008 Q1 Q1 2009 Q3 2009 Q3 2010 Q1 2011 Q3 2011 Q1 2012 Q1 2013 Q3 2013 Q1 2014

Q1 2010 Q1 2012 Q3

-500

-1000

Where did this brand new profitability at RFS come from with certainty if the 10-Q reports are a guide?

It must be noted that the spike started in Q3 2011, ie some 6 months BEFORE the “London whale” and ended actually towards the end of 2013, thus 18 months AFTER the “London whale”. On page 9 of the 10-Q of Q2 2012, the firm states: “In addition, management believes that the high production margins experienced in the second quarter of 2012 will not be sustainable over time” On page 13 of the 10-Q Q2 2012, the firm comments: “Mortgage fees and related income increased compared with the second the second quarter of 2011, driven largely by higher production revenue, reflecting wider margins, driven by market conditions and mix, and higher volumes, due to a favorable refinancing environment, including the impact of the Home Affordable Refinancing Programs (“HARP”), as well as higher net mortgage servicing revenue. Mortgage fees and related income increased compared with the first six months of 2011, driven largely by higher production revenue, as well as a favorable swing in the MSR risk management results (reflecting a gain of 6 million for the first six months of 2012, compared with a loss of $1.2 billion for the first six months of 2011).”

Thus this 100% profitability increase in an otherwise quite sleepy and constrained business did come from “market conditions and mix” but certainly NOT the level of rates or the economy overall. What was this “mix” other than the partnership that the IB and CIO had deployed since 2005 about RFS? It was actually Risk management on derivatives that was bringing this gain while it is routinely a cost…. As explained before, hedging the servicing fees generated by RFS was in itself a “basis risk” issue in full. On page 30 of the 10-Q Q2 2012, the bank draws the picture in the table below: “

154

155

As one can see the gains come from a huge increase in ‘production revenue’, and ‘risk management’ bringing up 50% of the increase each….Both are based on so-called financial ‘market conditions and mix’. How would the firm characterize this gain in derivatives trading related to RFS basis risks? One can notice here that the outstanding gains in ‘risk management’ result from ‘other changes in MSR asset fair value due to inputs or assumptions in model (a)’. Here we are! “MSR” means “Mortgage Servicing Rights”. The gain did come from the hedging strategies that CIO had deployed on behalf of RFS, in partnership with the IB, since 2005. As mentioned above the IB is related to that as much as CIO was through ‘FTP’ and ‘treasury and Securities’ and GCB: (a) represents the aggregate impact of changes in model inputs and assumptions such as cost to service, home prices, mortgage spreads, ancillary income and assumptions used to derive prepayments speeds, as well as changes to the valuation models themselves.” Mike Cavanagh had been on the front seat all those years on this specific matter. But Cavanagh will remain 100% silent on his Task Force Report of January 2013 on behalf of Dimon and the board of JpMorgan. As a comment the firm stated: “MSR risk management income was 4 million, compared with a loss of $1.2 billion in the prior year. The prior year MSR risk management loss included a $1.1 billion decrease in the fair value of the MSR asset for the estimated impact of increased servicing costs.”

The “fair value” had had issues….That was a $1 billion worth issue on a “basis risk”…..Now it remains to really understand what the ground was for this outstanding growth in the ‘net production revenue’. Was it really due to better or larger loan origination or was it also due to ‘financial market conditions’ like the ‘london whale event’? The table that follows illustrates some indirect reason as to why this ‘Net production revenue’ at RFS increased so much right in 2012 and 2013 only. One can see first that CIO usually had in fact a very limited input set at ‘other’ (bold green figures at the bottom of the following table below). Now one must wonder where the unrealized $10 billion recorded gains in the CIO portfolios were going (see April 5th email of Ina Drew to Jamie Dimon and the whole operating committee in the US Senate report exhibits where she mentioned already $8 billion) since they definitely did NOT show on behalf of ‘corporate/CIO’ either in Q1 or Q2 2012….

As suggested already, it may very well have been attributed to RFS in part but in the backstage behind the curtains, after the events when no-one would be looking other than the watchdogs who definitely monitored this.

The table below still shows that the ‘London Whale’ event timing (if not the event itself although Dimon made the link in august 2012) “cathartically” cured a problem that had plagued RFS all along 2011. That was a “fair value” issue already. This may be just one more calendar coincidence. Yes one more among many others in this scandal…. Most of the basic hedging (as opposed to the “residual macro risk” that was done by CIO) and valuing was done by the IB on behalf of RFS through the GCB and through the ‘FTP’ process that ‘treasury’

155

156 controlled with Cavanagh as chairman. The figures in red below show that a lasting issue plagued this ‘hedging’ item since 2010 at least.

There was an issue with ‘interest rates’ and ‘other model inputs’ from a pure ‘risk management’ prospective as well as with ‘other Fair Value adjustments’. The issue at stake amounted to some $3-5 billion it seems in total (see red figures circled below). It must be a coincidence in magnitudes with the $4-5 billion missing liquidity reserve for the “tranche book” of CIO alone that also existed since late 2010.

The firm balanced all that risk management loss with ‘valuation adjustments’ while regulators all scrutinized it then. That occurred all along the year 2011. Those ‘adjustments’ were ‘reported’ as ‘reserve releases’… That mattered to the OCC right? The OCC was embroiled “because of Jp Morgan” in the “robot signing scandal”. There was this ongoing concern about the CIO and its particular strategy since 2010. There was also the aggressive communication style of Dimon. There was the massive share buyback. There was the Basel III rule looming….There was the “stern” behavior of Drew after the MRA….The reserve releases were massive and could only balance crucially the final result as the sober $721 million net gain at RFS for 2011 as the 10-Q report shows. It is striking to see how next the ‘mortgage fees and related income’ jumped to an annualized $8bln pace in 2012 from a figure that averaged only $3 billion since 2009. Most of the increase comes from the reported ‘production revenue’ and one wonders how this is anything other than “basis risk” related gains…One can also notice and wonder about the $5-7 bln positive that ‘risk management’ brought in 2012 on ‘interest rates’ ( which barely changed actually) and ‘other’ Fair value modeling inputs’ that are NOT observable in the markets usually. This $3 bln (reported production revenue) and this other $5-7 bln gain in ‘risk modeling’ were also called ‘Fair value valuation adjustment’.

It was assumed here the root cause behind these gains at RFS were directly related to “risk management gain of the basis risk achieved on credit derivatives”.

How does the firm itself ‘explains’ in a way this sudden and quite fortunate increase in the ‘net production revenue’ through the different filings?

156

157

Page 13 of the Q2 2012 10-Q, “Mortgage fees and related income increased compared with the second quarter of 2011, driven largely by higher production revenue, reflecting wider margins, driven by market conditions and mix, and higher volumes, due to a favorable refinancing environment, including the impact of the Home Affordable Refinancing Programs (“HARP”)”. Page 226 of the Q2 2012 10-Q: “Net production revenue includes net gains or losses on originations and sales of prime and subprime mortgage loans, other production- related fees and losses related to the repurchase of previously-sold loans.”

The “reported figure” as such is a ‘net revenue’. As it turns out from the 2 extracts above, ‘production revenue’ is also driven by ‘market conditions’ and “mix”….It matters to clear the “mix” effect. Many ingredients were slipped by JpMorgan here as potential parts of this “mix”. What is for example the materiality of HARP influence in all this? The firm does not explain more though in August 9th 2012. In Q1 2012, the “HARP” program was NOT mentioned on page 10 of the report for the first quarter actually. Yet the jump in the ‘Net production Revenue’ was already quite significant in Q1 2012. This strong increase was therefore NOT due to “HARP”.

In the 2011 Annual report (published as of March 1st 2012), page 72, the firm explains” Mortgage fees and related income increased in 2010 compared with 2009, driven by higher mortgage production revenue, reflecting increased mortgage origination volumes in RFS and AM, and wider margins, particularly in RFS.” The steep decrease (see the table above) in the repurchase losses was thus a welcome event in 2011 from 2010. The firm thus justified the growth on the ‘production revenue’ from better conditions for ‘loan origination’ but this was NOT mentioned for 2012 in comparison to 2011. This ‘lower repurchase loss’ was thus NOT the cause behind this huge increase in the ‘production revenue’ for 2012. So this can only be due to ‘market conditions’, not ‘interest rate’ levels, not public program sponsoring, not business volume growth, not house price trends….

What is it then? Page 53 of the 2009 annual report, one can read “Mortgage production revenue declined from the prior year, reflecting an increase in estimated losses from the repurchase of previously-sold loans, offset partially by wider margins on new originations.”. Thus the firm had to report separately the ‘production revenue’ and the ‘repurchase losses’ but in fact the 2 figures are connected in the firm ‘reported’ modeled assessment anyway. It strongly feels that this ‘net production revenue’ increase is also driven by ‘unobservable inputs’. On page 61 of the 2010 Annual report of JPM, the same kind of statement is made “Mortgage production revenue declined from the prior year, reflecting an increase in estimated losses from the repurchase of previously-sold loans, offset partially by wider margins on new originations.” The firm does specify from 2009 onwards whether it includes the ‘repurchase losses’ by writing ‘total production revenue’ or ‘net production revenue’. Conclusion: the $3bln jump in ‘production revenue’ for 2012 is due to ‘estimated future costs’ related to ‘repurchase losses’ based upon other ‘model inputs unobservable in the markets’. But JPM would NOT make further clarification. It remains that this change in projected future costs (that were related to derivative hedging) would bring about $10 bln profits between Q4 2011 and Q4 2013 overall. Thus this $2.1 billion “explanation” from RFS was plain misleading in that it was NOT an opportunistic “release of reserves” but a lasting gain on the inherent “basis risk” of RFS that had been risk managed by CIO since 2005 actually.

c. Deferred tax “benefit” labeling mistake

More, one will see that in the “deferred Tax Benefit” line displayed on the “consolidated statement of cash flows” table, the bank reported a $1.47 Bln EXPENSE that it labeled mistakenly in August 2012 as a “benefit”. For those who are curious Jp Morgan had reported a real $444 million “deferred tax BENEFIT” for Q1 2012, an outstanding $1470 million “deferred tax EXPENSE” for Q2 2012 and a more nimble $755 million “deferred tax EXPENSE” for Q3 2012. If one tries to figure what the number for Q2 2012 could have been “in hindsight” using Q1 and Q3 figures only, one would arrive at (-444+755 divided by 2, or a $155 million deferred tax EXPENSE for Q2), one understands that the firm overestimated possibly the expense at the time by some $1.3 bln. This is as much that the firm may have been able to report through its earnings for Q1 and Q2 2012 “in hindsight” altogether without doing anything crazy. This $1.3 bln amount sounds like a gain that went

157

158 unreported at the time in August 2012. The labeling mistake, printing a “benefit” while it was an outstanding “expense” actually is unique in the 10 year record of reporting of Jp Morgan…. Was it another coincidence after the $25 Bln gain in tangible capital or else? Thus, here Jp Morgan apparently inflated its actual “deferred tax expense” by some $1.3 bln since it will revise down the “expense” in Q3 2012 while printing even more profits for the year 2012 going forward. The curious readers could simply redo the calculation done in the “consolidated statement of cash flows” for Q2 2012, along with Q1 2012, Q3 2012, Q4 2012. This check would show anyone that this “mistake” of the bank was unique. It appeared here in quite a timely fashion to make people believe that the bank could maintain its profit line “thanks” to an exceptional tax credit in Q2 2012. It also allowed to let people believe that indeed the firm had had to “release reserves” at RFS to balance all “this” loss. This conceptually indeed supported the misleading characterization that the bank had had to “release reserves” to balance “miraculously” the alleged “catastrophe of CIO”. It does not feel like a coincidence from that standpoint.

It matters to specify that this item among the consolidated cash-flows is NOT the ongoing routine provision that is made for taxes to be paid each quarter. It is a provision that is made in relation to projections made by the bank a bit subjectively. Below is the picture of both the original May 10th 2012 10-Q report for the Q1 2012. One will see that the bank here projected potentially a tax credit, ie lower profits somewhere for the bank :

As to the “restated” 10-Q report for Q1 2012, it will NOT be changed here because of the restatement itself :

158

159

It is quite interesting to see that this “price difference”, this “restatement” did NOT alter the cash-flows on a consolidated basis. One should ALSO notice that it had been an expected “benefit” in Q1 2011 too. There may have been a seasonality factor right?

How did it look like in Q2 2012 then? As the table below shows there seems to be a “seasonality” factor whereby in Q2 the bank tends to swing from a “tax credit” to a “tax provision”

WRONG label!

159

160

One can easily check that reporting gross mistake by looking at the 10-Q that had been issued in 2011 for Q2 2011 actually: here the bank provides its reporting rule in full

Thus Jp Morgan made a very gross misrepresentation about its provisions that here increased massively despite the possible “seasonality” factor. If 2011 was a guide, the swing from Q1 to Q2 had been of about +$900 million in 2011 in provision terms. It will be of about +$1.9 billion, ie an extra $1 billion provision actually at least. Thus the firm was far to be “plagued” by this “trading alleged loss” since its headline profit was unchanged (initially announced versus final) despite this extra provision here. Was there really such a seasonality factor? Not so much as the 10-Q report for Q3 2012 shows it so clearly…

160

161

One will notice above that the bank resumed its standard and correct labeling of the figures right after that period of August 2012. This suggests quite strongly that those reports are automated and that the forms have pre-programmed fields here. Thus this labeling “mistake” should NEVER have occurred anyway in Q2 2012. This shows also that one can definitely NOT conclude on any seasonality. Thus as stated before, the firm overstated its “tax provision” by some $1.3 billion specifically for Q2 2012 and this reporting mistake should be investigated seriously, as seriously as the “misses” of John Bellando…”Who” other than regulators themselves could have “reassured” Bellando that he could stop sending CIO P&L monthly reports to the authorities, for a couple of months like this? One sure thing here is that the bank still anticipated quite visibly some “extra profits” for 2012 no later than in July 2012 when it prepared the 10-Q for Q2 2012…..

d. Maiden Lane

As to the ‘maiden lane’ $545 million gain ‘reported in Q2 2012 in what looked like a ‘last hour helpful gain to save the face’ , it actually occurred in the first quarter of 2012, and had been long anticipated. The Federal Reserve really was well aware of that and yet saw JpMorgan postponing the gain from Q1 2012 to Q2 2012. This ‘Maiden Lane’ gain was ‘deferred’ artificially to Q2 2012 by senior management ‘in hindsight’ while the ALCO had had to “boost” the earnings by close to $5 billion from Q4 2011 to Q1 2012. All this occurred with the consent of the Federal Reserve here. More, This “maiden Lane” gain was underestimated most likely. Indeed the recovery on the Bear Stearns related note, looking at the federal reserve of New York website shows that the gain occurred well inside the first quarter of 2012, and was there for sure not in the early days the second quarter of 2012. The bank confirms it in its 10-Q report (page 180-181): “The Firm does not consolidate the LLC, as it does not have the power to direct the activities of the VIE that most significantly impact the VIE’s economic performance. In June 2012, the FRBNY loan was repaid in full and the Firm received a partial repayment of its loan. Therefore, during the three and six months ended June 30, 2012, JPMorgan Chase recognized a pretax gain of $545 million and $565 million, respectively, reflecting the expected recovery on the $1.15 billion subordinated loan plus contractual interest.”

It is said here : the gain of $545 million happened in Q2 2012 and this gain increased by only $20 million in Q1 2012. Thus the subordinated loan of JPM started being paid back. This note was NOT the first loss piece. The report of 2012 indicates that it was protected by an “equity piece”, being junior to the note of JPM. This note of JPM carried a 4.50% spread over the interest carried by the senior note. Thus by March 30th 2012, JPM and regulators knew that, first it was amortized now, second that it was well covered against future loss by the “equity piece”, third that the further amortization was quite predictable as the assets were unimpaired for most. This means that this note was worth par or close by the end of Q1 2012 already. The Fed NY website provides thus the reports that show that the JPM note was actually to be repaid very soon by April 3rd 2012. It was actually to be paid back by April 17th 2012 already further more. Thus this statement above of the firm in August 2012 is misleading. First, the gain was already “known” by April 3rd 2012 when the subordinated note of JPM was already being amortized. This further amortization was confirmed by April 17th 2012. Thus when the 10-Q report of May 10th 2012 was published this gain was there already but was NOT reported by JP Morgan. Second this gain had actually very little to do with CIO losses. It was NOT contemporaneous, OR, had it been correctly accounted for, it had to be allocated to Q1 2012 “in hindsight”, not Q2 2012.

It is worth reminding that the firm ‘restated’ $459 million for Q1 2012 but also deferred this gain of $545 million from Q1 to Q2 while it restated Q1. Thus this restatement should have been ‘contrasted’ with the fact that the firm did defer a known gain of a higher amount actually that the final restated profit figure. As to whether the firm then would have had a lower earnings profit for Q2 2012 as a result, it remains to be seen. For sure though, Dimon would have had to pick the $545 million from other parts like the $8Bln unrealized gains at

161

162

CIO investments, or the $2bln added provisions actually taken at RFS for Q2 2012, or the $1Bln remaining underestimation of the ‘reported’ fair value and the actual ‘measured fair value’, or maybe other unreported overcollateralization held within some other Federal Reserve run VIEs that were NOT consolidated as mentioned above, or the extra “tax deferred EXPENSE” of $1.3 billion….

e. Provisions : credit costs, CVA, DVA, litigation risks, CIO securities gains….

On top of that, there was no such ‘DVA’ gain but a ‘CVA’ gain of $755 million. The distinction made here bears on the fact that when one hears” DVA” one assumes that this gain was due to the widening of the credit spread of JpMorgan itself. In reality, as the bank specifies, “DVA” and “CVA” are managed altogether. Thus this gain was extracted in the activities managing actually, not the credit spread of JpMorgan alone, but the “counterparty risk” that Jp Morgan was facing, in other words the “basis risk” that lied underneath some $3 trillion of CDS exposures.

To secure that view one can read on page 142 of the annual report of JPM for the year 2011: “The CVA is based on the Firm’s AVG to a counterparty and the counterparty’s credit spread in the credit derivatives market. The primary components of changes in CVA are credit spreads, new deal activity or unwinds, and changes in the underlying market environment. The Firm believes that active risk management is essential to controlling the dynamic credit risk in the derivatives portfolio.“ One can next read (on page 81 in the 10-Q for Q2 2012) under ‘Credit portfolio management activities’ and ‘credit portfolio management derivatives’: “The credit derivatives used in Credit Portfolio Management activities do not qualify for hedge accounting under U.S. GAAP; these derivatives are reported at fair value, with gains and losses recognized in principal transactions revenue. In contrast, the loans and lending-related commitments being risk-managed are accounted for on an accrual basis. This asymmetry in accounting treatment, between loans and lending-related commitments and the credit derivatives used in credit portfolio management activities, causes earnings volatility that is not representative, in the Firm’s view, of the true changes in value of the Firm’s overall credit exposure.”

A table follows showing that the firm centralizes this credit derivative counterparty risk at the IB under the ‘credit portfolio’ activities. Yet this firm-wide management centralized at the IB desks plays a central role into how senior management applies its own massive adjustments as explained before especially in relation to the $3 trillion potential “basis risk” in notional amounts. Next JPM uses a reporting labeling ‘CVA’ as attached to ‘receivables’ and ‘DVA’ as attached to ‘payables’ from derivatives exposures. But the reality is that both figures come from hedging counterparty risks: when the gains come from ‘payables’, the firm strangely calls it ‘DVA’ while it really is a ‘CVA’ gain induced by a lower present value of what the bank owes to its counterparties.

This gain can show typically when the overall exposures have been ‘netted’ somewhat and when the gross notional exposures to counterparties have reduced overall. That gain $755 million gain should thus show as ‘CVA’, not ‘DVA’ as soon as one looks into the tables. But then (is this another “labeling mistake” like the “tax deferred” one, right?) the connection with the “tranche book” of CIO would have been straight, especially in the context of this “off-shoring”, also flagged by Dimon on April 5th 2012 to Drew as “our exotics credit wind down”….A more comprehensive table tracing the history of the overall CVA management on derivatives shows how exceptional the year 2012 was again performed right through the “London whale” scandal.

The scandal turned out again to be quite a profitable one event compared to other vintages. One can see that the impact on earnings was to be a $1.8 gain equally spread between H1 and H2. One can see in the red circle that this is most likely where the bank extracted the alleged $755 million so-called “DVA” gain. One can also note the figures in Q1 2012 and the quite outstanding $1.5 Bln “CVA” gain that translates certainly in part the “market improvements”. One can also see that on the so-called ‘DVA’ side there is an opposite result. But on a net basis there is $554 million gain that is completed in Q2 2012 by another $200 million gain in fact. The Q2 2012 alleged “DVA” gain of $755 million is really a one-off for what is really Counterparty risk management on credit derivatives overall also called CVA in practice.

One can also notice the significant reduction of the ‘Balance Sheet’ impact on the ‘asset’ side (down from $2.5 BLn to $1.7 Bln of impact). Those gains and reduction in size did not especially result from outstanding trading

162

163 activity from the ‘credit portfolio’ active monitoring of the counterparty risk at the firm-wide level as the lower part of this table shows.

Thus it seems that the ‘London Whale’ event sparked a $1.8 Bln gain in 2012 on an area, namely the ‘Counterparty Risk Management’ that is centralized firm-wide at the IB and used to be at regular loss in the past. This is confirmed in the following years although the check is limited as the firm will stop disclosing this information after 2013 it seems (the years 2008 and 2009 were available and displayed huge swings due to the crisis):

This second table shows thus a yearly profit of $1.8 Bln on “CVA” overall risk management for 2012, in markets that were pretty stable, while other years were a cost of about $500 million. This regular loss is showing in the P&L line of the ‘credit portfolio’ trading activity. One can also observe that the impact on the balance sheet changes radically after 2012. Thus the ‘London Whale’ is associated at JPM with a gain close to $2 bln on the “CVA” side and a deep rooted change in the way trading receivables and payables affect both assets and liabilities of the bank. Something definitely changed at JP on credit derivatives and produced a $2 bln gain as far as counterparty risk is concerned. This is not due to ‘credit spreads’ moving either way. The picture drawn by the firm, with this $755 million one-off DVA gain in Q2 2012 was quite far from reality.

Thus if one sums up the “provisions release” question the picture goes right at the opposite to what Dimon would let people believe between July and August 2012. Instead of a $2.1 billion sort of “release” from provisions, there was a net provision increase of about $1.5 to $1.6 billion actually, a massive unusual gain in “related income” at RFS, and a no less massive unexpected gain in CVA activities. How to arrive to this figure provision increase figfure? Well there was the loudly flagged “provision release” of some $500 million for RFS that was long planned and unrelated to the “tranche book” of CIO actually. There was this $1.8 billion inadvertent provision taken for “deferred tax” purpose. And there was another $332 million taken at

163

164

“Corporate”, the shell housing the CIO, taken for litigation purposes. One can find a very useful footnote on the matter on page 49 of the 10-Q for Q2 2012 for the line “non compensation expense”:

As one can see, this item is quite a significant contributor to the CIO performance although Dimon omitted to specify what is actually in the 10-Q in the (b) footnote here:

One can see that there was a $332 million increase of “litigation expense” for Q2 2012 alone. But actually the litigation expense over the first 2 quarters was increased by $2.8 billion. This impacted negatively the CIO performance by as much and this was NOT a trading loss. This is also as much provisions that were increased throughout the “London whale” event at Jp morgan at its CIO. Thus here again, the “explanation” of Jp Morgan was plain misleading in that the bank had NOT balanced the “trading loss” of CIO with provision releases at all. Instead, it displayed a red carpet “release” of $500 million at RFS, that in itself was completely predictable and should NOT have been mentioned as a balancing item anyway. But behind the scene the bank had taken a $1.8 billion tax provision and another $2.8 billion “litigation reserve” (on Q1 and Q2 in total), $300 million of which were taken right when the “contrite moment” had arrived….All this plagued the performance of Corporate by as much. Who knew that then? Who at Jp Morgan flagged the reality here?

If one wants to dig a bit further on this page one will notice that Corporate was specifically hit by these litigation reserves in Q1 and Q2 2012. This can be checked because the firm specifies which of those “expenses” were re-allocated to “other businesses”. Thus if one goes to page 33 of the 10-Q report for Q1 2012:

164

165

One can see that in Q1 2012, as opposed to Q1 2011, “corporate” had a hit of $2 billion ($3,328-$1,382) due to litigation reserves related to corporate itself. Was it the troubles with Philip Falcone LightSquared? Or was it the anticipation of something else to come in Q2 2012along the lines of the future “London Whale” scandal?

And in Q2 2012 the picture is as follows:

All this provisioning hitting Corporate came in Q1 2012, not Q2 2012. Thus the $332 million are very likely connected to the “London whale” event. It should be checked….

165

166

f. Restated numbers: another decoy…not a mistake for sure

The ultimate restatement of $660 million shows in ‘gross receivable’ and ‘gross payable’ fair values. It was attributed, as per the quite official thesis, to ‘traders at CIO setting wrong marks in the attempt to conceal part of the loss at CIO’…. It was pointed out before the inherent consistency of the “hit” that was impacting the “netting” process given the magnitudes at play between the gross $3 trillion and the ultimate “zero” to $100 billion outstanding exposure. The “story” did not hold water anyway.

The issue here is multiple: this was in fact NOT a ‘trading loss’ for the firm on CDS, and the CIO London traders’ price choices did NOT impact the ultimate reporting of fair value of the firm for derivatives as such. The firm says it actually but quietly …. See the table below comparing the figures. It has 3 sections: the stage 1, the stage 2 and the stage 3. Each display a key step in the “fair value measurement”:

The table above provides in summary how the alleged “price difference” impacted the “books and records” of the firm and triggered a mismarking and a restatement subsequently. One can see the 3 stages: the “ standard market price and routine collateral netting” stage, the “risk analysis and performance mathematical” attribution stage (fair value election, VaR consumption, RWA consumption, limit consumption etc…) and the ultimate “ALCO-earnings related” stage.

As one can see in the 10-Q reports for Q1 2012, the price difference allegedly changed the outputs of the bank risk and performance attribution systems. This is not a benign result at all…. The price difference of CIO allegedly disrupted the books and records of the bank Jp Morgan. Such is the blame. Such is the accusation of Dimon and all the top management of the bank. Provided indeed these price differences on a day could eprdure beyond the collateral nettings that were all done at the IB in a centralized manner every single day as far as the “tranche book” of CIO was concerned…. That never was the case by original design, described in the NBIA of 2006….

166

167

This accusation suggests that CIO was a sort of autonomous bank within the bank, ie CIO had its own VaR, its own RWA, its own limits, its own capital within the bank. It sounds like CIO had the very same status as the one of the IB within JpMorgan actually. This also suggests that CIO could have differing prices while facing the IB of Jp Morgan on almost every positions that the “tranche book” of CIO had. When one has ICE in mind, this “version” here feels like jpMorgan may well have been a sort of Ponzi scheme by allowing CIO and the IB have open trades together and keeping different prices within the consolidated balance sheet. This ALSO means in particular for the IG9 10yr that CIO and IB did NOT respect the contract that Jp Morgan had had with ICE since 2011 at least about granting ICE the role of “clearing agent”. It thus means that really not a single regulator did its job, ie the one they had since 1934. More they had ignored one crucial recommendation of theirs as a “lesson learnt from the last financial crisis” by ignoring the presence of ICE. Stunning!

What is a clearing agent like ICE? ICE is a “third party” that stepped in between say CIO-JpMorgan and the IB- JpMorgan every day on the IG9 10yr specifically so to impose the “market price” determined by ICE to both parties. The bank had signed up sometimes in 2010 already as far as CIO and the IB were concerned. More, the IB pressured the CIO staff to post the trades of CIO in the IB systems daily as early as possible so that the IB indeed would be able to meet the requirements of ICE and the associated contract of the “clearing agent”. Yes the IB also centralized all the trade bookings of CIO into ICE. This included the IG9 10year trades among many other index trades done on the “tranche book” of CIO. Thus really the IB did NOT do its routine day to day job at some point in time AFTER April 20th 2012 and this lad led to a surprising “price difference” as of March 30th 2012. All this makes total sense doesn’t it?! One may argue that, “maybe… who knows?” Yes, “who knows” aside from the regulators here? Jp Morgan and ICE had therefore a sort of loophole despite the daily reports and control checks that were heavily scrutinized by an army of controllers and the regulators alike. That was such a big loophole that it rendered the very function of ICE completely void of sense…

It remains that ICE was still netting notional amounts between the IB of JpMorgan and the CIO of Jp Morgan every day. And by the very thesis of Jp Morgan in July 2012 about this “CIO being a bank in the bank”, this means that- even with price differences- regulators could not have missed that the IB of Jp morgan and Jp Morgan as a whole had NOT the same net outstanding amounts in many credit indices like the IG9 10yr. There had to be a pretty big exposure “somewhere else at JpMorgan”. Then regulators could NOT have missed that this “particular hedging strategy” at CIO was responsible for most of this difference in notionbal amounts between the IB alone and the bank Jp Morgan as a whole firm. And this is quite intuitive here that the regulators would check that positions on an indentical CDS contract had the very same price within JpMorgan. Here we talk of tens of $billion if not many hundreds of $ billion in exposures on total…. The US Senate report accounts that the OCC monitored the IB performance daily. The OCC itself states on its website that it monitored closely the exposures of the big US banks on credit derivatives in particular. What was the OCC doing really? The OCC used to publish a quarterly report on its close supervision of the credit derivatives matter. Thus, if only to avoid the formation of a Ponzi Scheme within the walls of Jp morgan, it must have been that with or without the ICE contract, Jp Morgan and its regulators checked since 2007 that CIO and the IB had AT LEAST the very same price internally for all the trades where the CIO faced the IB in the markets. As such there was not a single chance that CIO and the IB could differ on indices like the IG9 10 yr once the routine collateral procedures had been run.

The April 2nd 2012 email between Hogan and Venkat about the 40% discrepancy about the 10% CSW risk measure testifies that the bank was aware of the problems related to price uncertainties, internal differences and they were on red alert then. Thus the focus on “potential price differences between CIO and the IB” was clearly there, quite visible and scrutinized, 100% contrary to what the bank pretends still in 2017 when it is said to have “missed” those price differences. What is therefore NOT correct in those 10-Q for Q1 2012 is that the price difference was anyway visible before the stage 1 of those reports was completed. It was seen by ICE. It was

167

168 seen by the OCC. It was seen by the IB collateral group managing the margin calls on behalf of the “client CIO”. It was seen by the top chiefs reporting to John Hogan. Really the question is : ” who at JpMorgan could have missed the price differences between CIO and the IB post the estimate P&L report was sent from CIO London staff?”

To be sure this does not mean that necessarily there was no violation of the books and records. Not any kind of price difference could be tolerated by regulators who also were watching them in angst. There could have been a price difference that was incorrectly set at the very start in CIO London. It is not as well to say that there should not have been a restatement. It remains that the bank should NOT have explained its restatement as shown above in any case. The 10-Q reports for Q1 2012 are thus BOTH wrong and misleading as far as the reporting of the restatement is concerned.

But this is NOT the only shocking aspect of those 10-Q reports where the “books and records” were restated “in hindsight”….One should notice that the ultimate ALCO stage figures are unchanged…The figures do not add up here as the OCC should have said. This is again a trick where only the “principal transaction” figures were impacted in fact… While obviously reserves were missing anyway if only to account for the price uncertainty on credit derivatives. Indeed one may argue endlessly upon the guilt of the “traders” here. It remains that such a tale was impossible to hold in liquid markets. One sure thing is that the “london whale” scandal had demonstrated that the credit indices and tranches were il-liquid. They should no bee “deemed” liquid any longer anyway.

By digging a bit deeper in the source figures of the restatement, one finds a disturbing truth. No doubt a change in price does influence the ‘collateral netting’ process across the whole chain. This is precisely – as pointed out earlier- why the price difference could not have lasted after the centralized collateral management had done at the IB day to day its mandated job for the “client CIO”. No, CIO was never allowed to be “a bank within the bank”. It has just been showed that such a tale was plain nonsensical given the facts of Ponzi scheme fears, ICE rules, bank organization.

One lasting argument here is that CIO by mandate was primarily protecting depositors “wisely”. This is NOT a bank mandate as such. It was under the shell of “Corporate” precisely because it could NOT be “a bank in a bank”. As a confirmed example of that one can see in the first section of the table above ‘see stage 1) that the price change through the restatement on ‘gross receivables’ (before adjustments) does influence of course the offsetting exposures on the same instrument using of course the same price change. Thus a loss of $297 million on receivables becomes larger to $367 million after netting. Similarly, a gain induced by this price change on ‘payables’ is reduced by the same amount after netting from $367 million down to $293 million. The total is $660 million, no doubt. This is mathematical and quite visible.

A price change impacted indeed all the trades being active across all the counterparties of CIO, including the IB of JPMorgan. One can figure out why this must happen this way by making up a simple example.

Let’s imagine that the firm has say a net $2 bln exposure on the IG9 10yr. It had sold protection for $1Bln initially. Its traded price then was 110 Bps. It had sold protection next for another $1Bln at say 95 Bps. The firm sold protection overall but traded twice actually with the same counterparty. Yet the firm has not tried to net the 2 trades into one with the counterparty (see tri-optima process for the experts). Thus the firm has 2 distinct open trades with the same counterparty. The market spreads had moved back to 100-101 Bps since then putting the first $1bln trade at a profit and the second $1bln trade at a loss. The first $1bln protection sold is recorded as a ‘receivable’ since the firm records a gain. The $1Bln protection sold will be recorded as a ‘payable’ since here this is where the firm records a loss. Thus the firm had ‘receivables’ for $1 Bln notional amounts and ‘payables’

168

169 for $1 Bln notional amounts. Every day, before any internal “fair value adjustment” was taken, the ISDA master agreement, the ICE rules, the US rules forced the firm to make a ‘mark to market’ ie confront its marks to the ones that its counterparties had on the same instrument and for every trade of the kind that is described here. The initial ( allegedly wrong) mark as of March 31st 2012 had been say at 100Bps. The gross receivables fair value reflected a 10Bp gain over the first $1bln notional while the gross receivable fair value for the second $1bln notional trade reflected a 5Bps loss. The ultimate ( allegedly right) price as of March 31st 2012 was 101Bps . There is about 1 Bp of difference here. It should have been “in hindsight” 101 Bps and not 100 Bps that was deemed too favorable in June 20112 only despite the close reconciliation reported by Allistair Webster on May 10th 2012. Thus the gross receivable figure reflected only a gain of 9bps (not 10 bps) for the first trade. Then, after ‘netting’ the ‘receivables’ and ‘payables’ for the same instrument with the same counterparty, matching the therefore the prices, the firm has also a larger loss on the ‘payable’ side of 6 bps instead of 5Bps (protection sold initially at 95bps and revalued at 101 Bps). This necessary full reconciliation occurred both in ‘gross receivables’ and in ‘associated netting’ operations for the very same instrument and the very same counterparty is why the ‘netting’ brings the loss up to -363 million and -297 in the table above in 2 parts.

This example above shows that both the ‘gross receivables’ and the ‘netting adjustment’ had to change when the ‘mark’ had been changed with every counterparty involved. This “netting” process was done with each counterparty separately so thanks to “legally enforceable netting agreements”

Now one has to look at the section 1 on the table first, and see the “net derivatives receivables” for example that is much lower than the total receivables. This is the natural effect of “netting” the “receivables” with the “payables” with each counterparty. Looking at stage 2 now, one sees where the risk models and performance attributions are displayed. This is, as per the “restatement theory” the moment when typically the bank would aggregate CIO positions and IB positions to indeed “consolidate” both entities into one firm-wide jp Morgan books and records official position. But here the figures are almost unchanged: there is just $70 million left to adjust. Thus CIO positions and IB positions were matching one each other at the end of March 2012 almost perfectly. This would explain why no counterparty complained then despite this huge price difference existing inside JpMorgan between the IB and CIO. Remember here that Artajo on the record of the call of the 23rd March 2012 reported to Pinto and Macris a $250 million difference based on just “one position”. And Pinto could not care less about that difference. He worried about the “very, very, very, very, serious accusations” of Drew actually. Once again this “tale” of July 2012 does not hold water in that it contradicts the very descriptions made by the bank suggesting that CIO may have been a “bank within the bank” or else a “hedge fund within the bank”.

This example above is not necessarily describing the way the firm itself opts to classify one trade or another trade, or the way the ‘netting’ adjustment is considered at JPM itself. This example showed how a netting process occurs and when in front of each counterparty having many trades at the same time with Jp Morgan IB and JpMorgan CIO. One can see that allegedly so the original “price difference” was still there in the “level2” bucket as far as the “receivables” are concerned for example (same picture in the “payables”). Now, as described before the “netting” now is NOT like $297 or $363$. It is only of $70 million ( either 120 000 - 119 030, or 118 353 – 118 283)!!?! Thus, once CIO and IB positions had been aggregated internally the “netting” change due to price differences was almost nil. This comment here only shows a fundamental inconsistency of JPM reports: on the one hand JPM states that the ‘netting adjustment’ was impacted by the restated price changes by say $300 million, and on the other hand JPM reports say that the ultimate netting adjustment firm-wide was almost unchanged at $70 million. The price differences were internal…

What is the difference between the 2 stages? In stage 1, CIO and IB positions were NOT aggregated firm-wide. In stage 2, CIO and IB positions were aggregated as the “level2-level3” breakdown testifies. The example above

169

170 showed that as much as one could have “believed” stage 1 with a $300 million or so issue, one cannot get “stage 2” where a meager $70 million impact is left on the change for the “netting” side. The example given above shows that the bank thesis is inconsistent in that the biggest disagreement on prices was actually internal to Jp Morgan, not external. One concludes that most of the price difference impact lied between CIO and the IB actually. What about the 90% of the positions held in the “tranche book” of CIO that had open trades with the IB in direct? That price difference was internal mostly and here this suggested a bit too much that JpMorgan is just another pretty gross Ponzi scheme. Indeed, one has to remember that CIO had large amounts of protection sold in the indices like IG9 10yr where the price difference existed. Artajo said that CIO differed from the IB prices by say 4-5 Bps, or about $250 million. Pinto knew it no later than the 23rd March 2012 along with Macris and Drew. He was NOT worried by that as such. How reckless he was if that difference was never cleared internally! Pinto was not reckless at all but super confident instead. He was because he knew that the CIO prices were overridden by his IB staff at the margin call stage ie at the very start of the netting process.

Therefore the price change could only exist IF the centralized collateral management located at the IB had let that difference last internally so. That was going counter to the most basic checks. This should just NEVER have occurred and was therefore done deliberately inside JpMorgan sometimes around the 18th April 2012 with a view to generate a fictitious mismarking for the end of March 2012. The internal price difference was blatant as of March 30th 2012 already as the March 23rd 2012 call with Pinto and Macris testifies. Yet no counterparty disputed CIO prices then. One then wonders why, in any case again, there was this short lived alleged collateral dispute around the 20th April 2012….

JPM did not alter therefore the ‘netting adjustment’ in a fashion that is consistent with its public statements about restating the figures in the books and records. This price difference was mostly internal anyway and had been manufactured deliberately within the IB around the 18th April 2012. One should recall the April 16th 2012 meeting with the OCC and the ensuing call from Drew to Artajo on April 17th 2012 “gui-dancing” Artajo to step out of the “bid-offer” quotes once again. That order of Drew was done in order to “alleviate the pressure on Hogan”….Drew had called Artajo here with a specific purpose that she exposed on the call…

Although the reasoning above may be subtle for people unfamiliar with collateral and netting operations, it is certainly the bread and butter of any big derivative player and every employee at the IB collateral management department that centralized all the CIO needs in terms of margin calls. The fundamental issue here, as the example made up above showed, is that while the ‘gross amount’ was changed due to the price change allegedly so, once risk systems had process their performance attribution between CIO and the IB… the ‘netting amount’ was almost unchanged. That makes no sense if a sole price change justified the restatement. Thus if one lines up the former argument and this one here, one gets to a strange observation…Firstly the “legally enforceable netting agreements” should have triggered disputes with CIO no later than March 14th 2012 as per the “March 5th & 6th order” that no doubt was executed. There was no such dispute. Secondly the ultimate “netting” done through CIO and IB altogether led to a minor adjustment of $70 million which indicates a benign “price difference” as far as the whole firm was concerned. This is consistent with the former point. Both points indicate that this price difference indeed was just “internal” and was adjusted normally. The restatement too was a fake in the form that the bank gave it “in hindsight” in August 2012…

Was Jp morgan running a Ponzi scheme actually with its CIO under the watchful eyes of regulators since 2006 as the NBIA of 2006 would testify that CIO “should” but did not “have to” pick the IB prices on a day to day basis? As of March 23rd 2012, Artajo elevated the price difference to Pinto and Macris. Macris stayed mute. Pinto was assertive that “nobody had mismarked” the books. Pinto was even ironic about the claim of CIO. They had just no dispute at the IB that Pinto had heard of while managing the margin calls of CIO every day then. CIO had notoriously had a leeway that was adjusted daily at the IB for margin calls purposes. Indeed one

170

171 has to remember that the NBIA of 2006 (see the US Senate report exhibits) stated that CIO “should”’ use the IB marks….But CIO just NEVER used the IB marks all those years notoriously so as US Senate report itself confirmed officially on March 2013 the 15th…

Something is wrong here -for sure- irrespective of whether JPM had the rules described in the example above or other rules (like protection sold is a receivable and protection purchased is a payable, or like the netting adjustment is realized P&L or else…). As much as the prices were not what they should have been at the firm- wide level, this reporting self-inconsistency shows that the issue was not with the CIO marks but with the ones that were in the collateral accounts held on behalf of CIO, at a stage of the valuation that occurred after CIO had communicated its own marks to the IB staff. Indeed, if the CIO marks alone had been wrong and used straightforwardly in collateral management, the firm should have altered accordingly both the ‘gross receivables’ and the ‘netting adjustment’ associated with receivables accordingly at every stage. At the first stage it had to trigger a dispute no later than the 14th March 2012. No such dispute was ever reported…. Had it been internal truly, the IB would have complained internally against CIO no later than the 23rd March 2012. The US Senate report exhibits provide an email of Mark Demo dated April 20th 2012 that explains that they at IB- collateral had observed a somewhat $500 million difference by the end of March 2012. This difference had been elevated then already by Demo or his team to “supervisors”…Yet there was no mismarking. The irony of Pinto on the call with Macris and Artajo shows that was not a single chance for that to happen. The reason is obvious CIO is NOT a contender of the IB in the mark to market process of the bank as soon as the collateral stage starts.

Had this alleged price difference really lasted, this would have changed the stage 3 ALCO figures too. But this would have meant that IB and CIO were altogether running a Ponzi Scheme inside JpMorgan, not a hidden hedge fund…And in fact the bank did not alter the ALCO figures…..Either the ‘netting adjustment’ changed and this was because of the price change. Then the firm had to consistently impact all its tables automatically Or the netting adjustment did not change actually as the Table 2 shows and the $660 million adjustment had no reason to show in the ‘gross receivables’ figures anyway (if the bank is to be believed in its suggestions). In the former case, it could have been indeed due to incorrect CIO prices. But then the whole fair value chain would have corrected it before stage 1 had been completed. And it was NOT the case. In the latter case, that the netting adjustments were unchanged, this strongly suggests that this ‘mismarking’ had its roots in the reserving process, not the front office marking process of CIO London.

Therefore the change for restatement as reported in the sole ‘gross receivables’ was inappropriate and misleading. The same remark applies to ‘payable’ and their associated ‘netting adjustment’. Thus the marks that were restated were NOT the CIO ones but certainly the ones that the firm had used on behalf of CIO away from CIO sight on May 10th 2012. Therefore the ultimate 10-Q for Q1 2012 was false in August 9th 2012 too: either CIO marks were not at fault for the restatement and the mismarking was elsewhere (which indeed would explain why the ‘netting adjustment remained unchanged in stage 2 and all the fair values down the line too through the whole firm valuation process applied to credit derivatives), or CIO marks were responsible and the netting adjustment should have had a different value all along with all the subsequent fair values related to the valuation of the credit derivative bucket…Whatever the case the 10-Q report for the Q1 2012 was false be that as of May 10th 2012 or as of August 9th 2012 “in hindsight”.

One will not see really how this $660 million will be turned down by the bank to reach a final $459 million ultimate restatement of its earnings since the bank does not flag the audit track in its 10-Q reports within the ‘unaudited’ ‘Non-Interest Income’. This is quite logical in fact if one understands that all the subsequent adjustments made by risk management for CVA-DVA-unobservable inputs and reserves is based upon one central price database that is used independently through the GCB-WSS-FTP independently of whatever prices

171

172 came from CIO London Front Office or counterparties. To be sure, the firm’s risk management certainly uses those price sources like the CIO London one at one early stage but it quickly applies its own ultimate prices next. Thus, any early price change at the ‘gross’ fair value stage is NOT impacting the ultimate Fair Value figures for the firm by design.

Thus this restatement ran contrary to the firm’s own process logic and was actually misleadingly reported by JPM in august 2012. One may say:” Come on this is just $660 million! Granted the way the restatement was made was misleading but it weighs nothing against the $5.8 billion loss in the firm” That is quite true in any event and operated just as another distraction from the main economic impact of the ‘London Whale ‘ event that was something like a $50-60 billion unique historic gain “net of all related costs” that happened timely so as planned since 2010.

5- Conclusion: the bank made huge gains, acted on purpose all along and disguised them as much as possible. This was a “marionette” show in a teapot at the end of the day…

The year 2012 vs the others: it was a bonanza year not to be repeated before long….

The bank in 2012 realized a long awaited $25 to $75 bln quite tangible and predictable gain in “hard capital”. The ultimate gain would be spread over time till the end of 2014. Such achievement normally requires at least years of track record in growing predictable profits to make such an achievement. Dimon can be praised for that technically speaking. After all the failed attempts from regulators to provide a sufficient capital lifeline to the JpMorgan brand since 1999, at last Dimon “succeeded” in July 2012. Dimon completed his plan right through this “London Whale” scandal to secure the “tangible capital” basis on behalf of the name “Jp Morgan”. Overall the “off-shoring & internal collapse” of synthetic tranche positions across the banking group Jp Morgan was sensible, wanted and would bring up an historic improvement of the balance sheet. That was NOT a “mistake”, NOT a “loss”, NOT an “incident” at all. “That” was an outstanding and rare success. “That” has been “closely managed” in a fashion that was quite the opposite to what all the many misleading characterizations suggested since Q2 2012 even if the original scenario of 2010 where “one marionette trader” was to bear a well manufactured blame would not be the ultimate outcome. “That” did not have to become a trading scandal.

Initial descriptions and later characterizations “in hindsight”: myths and reality

The positions of CIO in the “tranche book” were well balanced but huge still as the former parts described. The drift of the “forward spreads” was a plague. Yet this nasty drift was well balanced by other exposures in a quite rational way within the “tranche book” of CIO still. The net resulting loss in January and February 2012 was mostly due to trading costs, induced 99% by Drew’s express and repeated orders of the time. She was perfectly aware of the risks that she was building in the book here. Iksil would warn all along accurately, timely and truthfully. Iksil next would express his refusal to keep trading as of March 1st 2012 at the risk of being fired. Iksil had warned in January 2012 already. Macris and Drew will react by demoting Artajo first, by calling Bacon in the loop next and by initiating the March 5th and 6th 2012 orders altering unquestionably the CIO London valuation process. This order here had to trigger an amendment of the NBIA of 2006 as they knew it as original designers of the NBIA in 2006…. But the “post implementation review” had never been even started…

The bank was responsible for that in full as much as the regulators were since 2007…By March 2nd 2012 CIO had surrendered but the easy preliminary “transfer from CIO to the IB” would not happen… Not yet anyway…The March 12th 2012 meeting with Bacon first caused all the winning positions to lose all at the same

172

173 moment but in a shadowy fashion. Had Macris and Drew anticipated that already as of March 2nd 2012, then the motivation for their March 5th and 6th orders becomes crystal clear now that they had “surrendered”: they meant to claim that the future losses would ONLY come from Bacon’s “leaks” or “traders judgment mistakes” creating a free lunch for themselves either at the IB or at CIO or both. The March 1st 2012 email of Macris where the CIO London chief confessed his concern about his ability to “defend the positions” in the future is a crucial clue in that regard. That series of well synchronized leaks could NOT be due to external market players anyway since they could not know ALL the positions in such details. To be precise, not a single IB trader at Jp Morgan could organize these leaks either. Yet the information came from within JpMorgan or from regulators (who knows right?).

That conjunction of events (CIO business Review on February 29th 2012, then March 1st, March 5th, march 6th, march 12th, ensuing losses and march 23rd theatrical elevations) could NOT be a coincidence since just all the historical relationships in market prices started being broken then. The articles of April 6th 2012 increased the burgeoning loss but not so much as most market players did not believe in this quite superficial and counterfactual “london whale” legend, including most among the IB traders of Jp Morgan itself (see Cedric Lespiau-SG or August Lund-GS descriptions and others at the IB- see the March 23rd 2012 between Stephan and Artajo about Ari Weischman for complete information). Yes the IB traders were involved in “hedging the CIO position increases”. Yes they would remain silent later on as the firm threatened them with litigation and claw- backs if they expressed their point of view, whatever it was.

But some medias would keep making allegations cumulating the nicknames as if they could overwrite the facts forever. Pushing the IG9 10yr skew to zero in those “no man’s land” markets was unrealistic for most market players, and just all the regulators….Yet it would happen.

Thus, as of May 9th 2012, the IG9 10yr skew was not expected to go at zero. And therefore the immediate gains that could be obtained from this “London whale” operation for the bank itself were NOT maximized. The markets on the whole, apart from say Weinstein, Hubbard, Vilani, Goldenberg, Feldstein and Guy America at Jp Morgan, did not “buy” this tale so much about “Jp Morgan vs Jp Morgan”. Why would they actually step in “family affairs” where JpMorgan made a “mockery” of itself (Macris to Pinto on March 23rd 2012)? This is only the May 10th 2012 statements of Dimon that would cause the ultimate and massive distortion in the markets, ie the distortion that would bring the IG9 10yr skew at zero fast on the follow. These statements alone caused at least 2 thirds of the total $6 billion loss at CIO. This brand new situation at last maximized the tangible immediate gains for the bank itself. What a unique windfall that financial history had disproved for decades!

As Drew would write in CAPITAL letters to herself on the company’s email system, Dimon had weighed the risk and rewards here over the previous week. That was intentional, ie not out of control. Then the markets read it straight through the tea leaves: CIO was gone for good and in shame. This was the “go ahead! The ig9 10yr skew can go to zero”….And this is right after that that the final stage of the plan was reached, ie when the IG9 10yr skew was at zero at last. Then the bank let the word go in June 2102 that it had “disposed” of the CIO positions. Yet the bank did NOT trade much in the markets if at all. It simply transferred the “tranche book” of CIO into IB books as a preliminary step. It was so simple!

This preliminary transfer of the “tranche book” of CIO to the IB could have been done on January 4th 2012. It could have been done on October 1st 2010 actually, right after the September 2010 presentation of Dimon. To be sure it was do-able at any point in time since 2009. It was officially what Dimon wanted to happen since September 2010. It was just an internal transfer that gave way to an internal collapse at the end of the day, through a third-party maybe, aside from minor “collateral damages” especially in human terms. For this mostly internal operation, the IG9 10 yr skew price through the easy preliminary transfer would have determined the

173

174 immediate gain, or the immediate provisions that the bank would have had to record for what should have always been called an “off-shoring and subsequent collapse” of “exotic credit derivatives”. That is what Dimon had organized since 2010. Either way, over time it was quite a predictable and tangible gain in capital : it was worth $50-70 billion anyway net of “collateral costs”.

Rather than executing this “wind down” plan in late 2010 and extract a net $15 billion generation upfront gain of “hard capital” the bank will go sideways. Another mistake no doubt….Then the “London Whale” scandal will burgeon in Q1 2012 and blossom on April 2012. It would be placated then in 2012 and until 2017 that the positions of CIO were “flawed” by some media again and again…. But the bank would not unwind them in the markets anyway…. It would be also tried again and again to discredit Iksil’s version whatever it was….But the tale was pretty blurred anyway. The media were soon short of nicknames here in July 2012 and people started wondering aloud in the press… What was going on?

The positions at CIO would have mean-reverted in 2013 despite all this gesture as the former part showed as well. This also was quite predictable anyway for any experienced market player even in the middle of the turmoil that the bank contrite statements fueled clearly between April and August 2012.

The VAR “incident” and the mischaracterized balances of gains versus losses leave little doubt on senior management candor

There is very little doubt that the top management acted with the “tranche book” of CIO in order to maximize the quarterly earnings all along between 2006 and 2012 (see the “maximize P&L” of Drew on Jan 10th 2012- see also the April 5th email chain between Drew and Dimon). As such it does NOT prove that the bank top executive definitely manipulated the markets for that outcome to occur. Yet one can read usefully the exhibit called “VaR history” on this website. It tells a story that is pretty consistent over time: all the levers were activated around the VaR of the firm as a whole for this “tranche book” of CIO to “maximize P&L” at the firm- wide level actually. That was not “just greed” or hubris or anything like that….At least not on the side of the bank… The “VaR history” thus would be showing that the CIO “tranche book” turned out since 2008 to simply help the IB maximize its gains quarter after quarter. That had not been the original intent back in 2006. But the year 2007 would mark the turnaround on the matter. The reason was quite strategic and stretched way beyond the paper walls of the “fortress balance sheet”.

The ever deeper lack of liquidity in the CDS markets was the cause that pushed the bank to use this “CIO tranche” book until its very last day more and more exclusively to maximize the upcoming profits of the bank. This is what explains in full why the top of the firm, from New York based Hogan’s teams, no later than the 27th January 2012 likely tweaked some “cross markets” correlation figures to minimize the VaR of the firm. The watchdogs were watching that out and would let that happen. This tweak critically allowed Drew to maintain her orders to Artajo that forced Iksil to trade and trade again and again. The US Senate report and its exhibits show it if one starts focusing closely on the emails coming from Keith Stephan and Irv Goldman of the 20th January 2012.

Had the very top of the bank, involving Dimon then, not minimized the firm-wide VaR so artificially then, none of the “trading strategy” of Drew, as wanted by Drew repeatedly despite Iksil’s alerts, would have seen the light of day in January 2012, February 2012 and March 2102. There would have therefore been no ground for the “London Whale” legend or myth to take hold in theory. The only ones at the bank who wanted Iksil to trade on and on then were the top executives of Jp Morgan including at least, Dimon, Drew, Macris, Hogan and Braunstein…..This strategy of theirs here, to have Iksil trade on and on in notoriously il-liquid markets for quite concentrated and visible positions, had really started in February-March 2011, ie one full year before the

174

175 scandal….Evidence of that exists…(see the US Senate report for the stress test violation reports dated end of January and February 2011 among other documents that should be made public…). Regulators then would turn blind, inactive and would never try to meet with Iksil the “new MD” on their “SCP”….

The top bank chiefs had long wanted this “trading pressure from CIO” to show in the markets. Their motivation to place this “human screen” as a ‘fall guy’ was not new for them. It dated back from 2008 and 2009 in fact when they had made serious mistakes knowingly so, witness the OCC MRA, the FCA “supervision” letter and the contemporaneous correspondence with the Federal Reserve… that all landed at jp Morgan in late 2010. They would miss here a unique opportunity to officially set this “tranche book” in run-off mode from December 2010 onwards. They did not take the opportunity and that was just one more ‘mistake’…. Time was running out in early 2011 already for them….They would hook a “new MD” for the fall to come…

The regulators, despite their ongoing “close and continuous supervision” since late 2010, were maybe “surprised” to read about the positions in the media in April 2012. But the fact is that they were very “familiar” with the matter since 2007. The top of the bank had concealed its intents even through the July-August 2012 statements admittedly so if the “settlements” of 2013 are a guide. But should the admissions be trusted? No. A lot more should be disclosed on the matter as Carl Levin pointed out.

One can notice the multiple mischaracterizations present in the 10-Q reports and the broad denial that continues in 2017. One can remember the quite artificial $6 bln “trading loss on credit derivatives” set at the ALCO stage only. This loss showed up in the “principal transaction” line but in reality it was planted here by the senior executives AFTER market prices had been selected and reconciled, AFTER routine collateral tasks had been performed, AFTER mathematical risk models had been re-allocating performance between businesses, AFTER the strategic “fair value election” options had been scrutinized by ALCO itself and risk management….Regulators could NOT have missed that while they were all suing “CIO London traders” only.

One must assume here that regulators backed this really deviant characterization. One can ALSO see how misleading the statements about “provision” releases were since provisions were actually increased (not released overall) in Q2 2102 to the tune of a net $1.5 bln as explained before (take litigation and deferred tax expense along with actual credit cost provisions). One could also remember that RFS (Retail Finance Services that normally “make loans to customers”, ie “basic banking stuff” for the man on the street) actually would cumulate a $10 bln extra gain that has little rational explanation other than a suddenly massively improved risk management skill on derivatives hedges based on “basis risk” structurally….Yet this sudden emergence of “skill” vanishes in late 2014 along with some un-audited, un-consolidated in full, off-balance sheet derivatives “basis” exposures that are NOT carried at Fair Value since there might not be “legally enforceable netting agreement” in place…The latter description would perfectly fit with the legacy skew exposures of the bank, based on “identical underlying”, that had been “off-shored” (FCA words here for the operation of externalization) by the end of the first half of 2012.

This must be a perfect coincidence in timing between two vastly unrelated businesses, at least on the surface if the bank “explanation” can be trusted…. In terms of coincidental events, one could also remember the unique and weird labeling mistake about those “deferred tax” benefits that were provisions actually inflated by some $1.3 billion ‘in hindsight” for the sole second quarter of 2012. Was it another ‘mistake’ here that went unreported as such?…. One should also recall that “Maiden Lane” coincidental calendar shift. Indeed, absent this quite artificial postponement of the quite predictable Q1 2012 gain in “maiden Lane” of $550 million into Q2 2012, there would have been just NO restatement of $550 million for Q1 2012. It is all the more surprising to see that this calendar tweak pop up here in July 2012 as the Federal Reserve was directly involved in this “maiden lane” deal since 2008 and watching…. One could see on the Federal Reserve Website itself that the

175

176 path to expiry of the “Maiden Lane” deal in question was indeed announcing quite a predictable gain for Jp Morgan for Q1 2012, like in late 2011 already.

One should at last try to get to the bottom of the quite technical way the firm actually “restated” through the Amended 10-Q report for the first quarter of 2012 and used those “restated” figures in the 10-Q report for the second quarter of 2012. With a little use of common sense one would see that the “netting” figures are inconsistent with the thesis deployed by the bank between stage1, stage2 and stage3. Therefore the August 9th 2012 10-Q reports were NOT to be trusted anyway. As to the “price difference” itself that was invoked as the root cause for the restatement, it is plain wrong as here Jp Morgan denies the existence of a control standard that Jp Morgan itself imposed to the banking industry in 1993 with the blessing of no less than the SEC, the OCC and the famous Paul Volcker. The alleged persistence of this price difference, as stated by the bank and later supported by the investigation reports, also strongly suggests that Jp Morgan was run as a gross Ponzi Scheme being here much less subtle than the one of Bernard Madoff… Forget then about the CIO being a sort of “hidden hedge fund”. The bank’s own “explanation” for the restatement itself leads to conclude that it was just a gross Ponzi scheme sitting between CIO and the IB. Unbelievable right? Yes, indeed. And yet, a few infamous nicknames and few contrite statements would create enough smoking mirrors to make believe….

This was indeed just a “tempest in a teapot” but one that the bank wanted to have in some respect

The mechanism at play since 2005 is crystal clear…. Drew is central and she reports to Dimon straight. Regulators watch all the developments quite closely all along if only because the Sarbanes-Oxley laws (2003- inherited from the ENRON scandal among other things like the close involvement of JpMorgan then) mandated them to do so. CIO was created in 2005 with a purpose that was directly related to the unique billion of intangible capital that had emerged through the “merger” between JpMorgan-Chase and BankOne in early 2004. Granted “Jamie” is good looking and inspires “Trust” then. But regulators could not have missed that intangible billion amount since they had “approved” this quite innovative merger on banks ledgers. The NBIA is finalized in 2006 but its “post-implementation” form is never filled in an apparent breach of the Sarbanes-Oxley guidance on the matter…To be sure, the guidance was that a “post implementation” review had to be performed and “someone” had to be in charge of that. At Jp Morgan, for this “tranche book” of CIO, there was not even a “someone” that will have been tasked to start this “post-implementation” review…. And regulators missed that right? Still in 2007 CIO and Jp Morgan will put enormous trades through this “New Business Initiative” that had been “Approved” through the NBIA at least internally at Jp Morgan. And those huge trades will remain actively “managed” by the top of the firm until June 2012 without any “post-implementation” review being started in any way. And regulators asked questions here since 2008 (see the US Senate report). Here it is really worth going back to 2002 and 2003 to retrieve the “spirit” of the Sarbanes-Oxley legislation….This “version” like “we were unaware” from the regulators does not hold water.

The OCC in 2012 despite its very gross and broad ignorance of this “CIO tranche book name” (US Senate report) had to admit that it knew of this “particular strategy” since 2007. The enormous trades here since 2007 were piloted by Dimon and Drew as David Olson based in CIO New York testified independently so through the media. The top executives had recruited specifically one big “trader” to execute in the markets on their behalf. That was Javier Martin-Artajo and no-one else. The massive Subprime trades among others had occurred but only after the fresh recruitment of Artajo in February 2007. The CIO-London estimate P&L report key developments, under Artajo’s close supervision too, were made in September 2007 at Drew’s personal desires under direct CFO watch (Joe Bonocore then- the emails do exist to prove it). The “unwind at no cost” order in 2008 was done as per Drew’s instruction through Artajo on a day to day basis. The “re-load” in 2009 and ensuing “unwind at profits” orders likewise were also issued by Drew under Drew’s close watch. The removal

176

177 of the “$300 million cushion” in late 2009 was done as per decision of CFO-New-York going AGAINST “traders” (ie Iksil and Buraya) repeated requests. That was NOT the ‘traders’ job here (see Artajo on the matter).

The new ‘land the plane’ strategy in 2010 was executed under close supervision of Macris here helped by Kalimtgis and Stephan for Drew. A daily “deleveraging” report was sent to Drew on the matter then. The “dividend trade” in late 2010 was deployed and unwound soon after as per Drew’s instructions through Artajo who wanted it as a ‘tactical’ strategy then. The “MD” promotion of Iksil was done soon after with Drew’s blessing and Dimon personal approval in November 2010. The removal of the “$100 million cushion” again in December 2010 was done as per decision of CFO going AGAIN AGAINST “traders” (ie Iksil, Grout and Buraya) request. Indeed, that was NOT ‘traders’ job at CIO to worry about ‘unwind costs’ or reserves for this “’tranche book” even though they were to unwind day after day…and report to the higher ups faithfully so..

Next, surely by coincidence, there would be this Drew’s stern reaction in front of the OCC’s MRA and these ‘stress test’ misleading reports that were sent under Drew’s direct watch in January 2011. All this gesture sounds quite hyprocit. The trading activity on the “tranche book” of CIO was close to nil due to il-liquidity. Drew’s massive sale of treasury in Q1 2011 (backed by Dimon all along) had caused the CIO’s stress test limit violation at least in great part. The events here occurred unbeknownst to the “freshly promoted MD” Iksil. Following a key alert of Iksil in March 2011 face to face with Drew about il-liquidity and unwind costs, Drew finalized the “split approval” in June 2011 (see Strategy 27 creation). The “tranche book” was officially in “run-off” mode then, expected to expire soon. BUT Drew next would order to grow the forward spread trades in July 2011 onwards through Artajo. Jp Morgan had announced internally a plan to eliminate its CIO and the “tranche book” of course (Artajo knew here BUT Iksil “the fresh MD” was NOT in the loop). This radical change occurred at some point in time in the course of the summer 2011. One may wonder whether the unusual sale of treasury holdings around February 2011 - causing the stress limit violation as per Artajo-- was not a precursory event of that outcome for the planned “scandalous” extinction of CIO …..

Next there were Drew’s paradoxical orders of December 2011 9th in this context of run-off mode (“get net long risk, cover HY risks”). There was this last-minute early year end valuation that was “officially” meant to close the books as of December 15th 2011 for the “CIO tranche book”. By another coincidence Drew would have to address Federal Reserve CCAR-related requests targeting the “unwind costs” for the “tranche book” specifically. The FED requests had been officially sent to “Dimon and Al” between 22nd and 27th December 2011. A second year end valuation will be performed on the “tranche book” as of the 31st December 2011. Why was this second valuation needed? Surely it is because there was NO problem with the first one right? There already was some $300 million of “price differences” inside Jp Morgan about the “tranche book of CIO” between CIO prices and IB prices. It was reconciled on the day but it popped up the day after steadily…

Regulators were really involved in late 2011. Against all odds, the US Senate report disclosed Drew’s “maximize P&L” order of January 10th 2012. But NONE of any public reports would disclose Drew’s anger at increased trading costs on January 18th 2012. Still Drew will order to “cover Kodak loss” on Jan 20th to meet “regulators’ expectations”. Following many new warnings of Iksil about Drew’s recent instructions and price uncertainties, Drew will conclude “I am not worried” on Feb 3rd. Still she was demoting Artajo right then in the backyard because Macris WAS worried (see Iksil’s alerts on January 30th and 31st 2012). But the “freshly promoted MD” Iksil was NOT told by either Kalimtgis, Stephan, Macris, Artajo, Drew, or Human resources of this fresh demotion of Artajo. Faced with a renewed warning of Iksil on February 7th or 8th 2012, Macris will order Artajo to “take the plane to NY” for Drew’s orders to be completed. Artajo on the face of it complied and got a very explicit help from Hogan the firm-wide CRO of the time.

177

178

Then there is the Zubrow’s “Volcker rule” 68-page long letter sent on February 13th 2012 to just ALL the authorities involved with Drew closely involved too (see Us Senate Report). The “tranche book” of CIO was un-mistakenly referred to in this letter. No one ever heard that ANY of authorities receiving this 68-page long letter EVER asked a question about this book or the coming fate of this book. They knew. It was already officially to be DISMANTLED or “TAKEN DOWN”. However, right then, the top executives of Jp Morgan granted unlimited BUT temporary limit extensions for the trades of Drew to be completed, dismissing thus in full Iksil’s alerts and advices on the matter. One wonders at this stage… It feels like all the senior management wanted Iksil to trade like a mad man right then, right? There will be next Drew’s and Dimon’s CIO business review on February 29th 2012. Were regulators “unaware” of those recent developments? Iksil will report on March 1st 2012 the call that he just had had with Gabriel Roberts (CITIGROUP) and will state that he would refuse to trade in the future even if that meant that he would be fired. Macris heard it after Artajo did. Then many things happened….

Macris called Bacon in emergency no later than March 2nd 2012, concerned by “our” inability to “defend the positions” in P&L terms (see March 1st 2012 email of Macris to Artajo). Drew sent new instructions to the “now demoted” Artajo on March 5th and 6th 2012 (see Iksil’s slides of late February 2012 projecting for March 2012 only a loss ranging between $300 and $800 million and yet Iksil’s refusal to keep trading in March 2012-)….. There will be the key March 12th 2012 meeting with Ashley bacon, and ensuing Bacon’s decision on March 14th 2012 to “offshore” the CIO “tranche book”… As per Bacon then, there were many “synergies” with the IB books BUT “regulators’ approval was required in the first place”….So, regulators were queried, right? But the preliminary easy transfer “from CIO to the IB” of the “tranche book should have been done then, without the need for the regulators approval since Bacon had taken over, right? Who would imagine that Bacon here did not make a roundtrip among regulators starting no later than the 14th March 2012? Who would imagine that Bacon did NOT have the backing of Dimon here in postponing the easy preliminary transfer while CIO had surrendered already?

The top executives will then move pieces in a well concerted way with some “noise” emanating from Drew actually…. This is probably when it started “running out of control” as Dimon would state in 2017: the well prepared “fall guy” would not be Iksil alone….Drew starts freaking really as per March 16th 2012. March 19th 2012 compliance is alerted by Iksil upon Artajo’s express order (see again Iksil’s alerts on the previous week)….But Iksil does NOT blame the IB so much…. March 20th 2012 DRPC Board meeting with Drew picking March 6th 2012 as last valuation date peculiarly so and December 2011 for the other risk metrics….. But Iksil loudly warns again on the March 20th 2012 of “more losses to come”…. Macris is “scared” and Artajo feels very down…. March 22nd 2012: this is Drew’s official “freaking” moment (see again Iksil’s alerts on March 20th, 21st , 22nd 2012 estimate P&L report announcing the end of the trading for good through Grout)…Right after Iksil alerted again on March 23rd 2012 first time in the morning of further $300 to $600 million losses to come by the end of the month of March 2012, and fought CIO’s managements initiatives to keep trading even away from him, see Drew’s elevation going “all the way up” : “Ina is not the most stable person in the bank” Pinto the JPM UK CEO would tell Macris…. Drew ordered then to “put the phones down the time for us to have the new RWA figures”….

Drew would allege later that she then “discovered” trades that she had not heard of before….But at the time of her alleged “discovery” Drew will unwind NOT A SINGLE trade until Bacon and O’Rehilly are sent by Dimon to take over on April 27th 2012…… even then, the tandem sent by Dimon shall ONLY remove the most liquid positions and quite slowly in fact… Was it really running “out of control”? There was good cause for that motionless behavior all along April 2012 as all the top executives being in charge knew very well what they were doing here since the very start of the year 2012…The easy transfer still was not to occur soon. Things remained under full control despite the VaR massive breach of January, despite the demotion of Artajo

178

179 and Iksil’s alerts in February 2012, despite Drew’s gesture of the 23rd March 2012. See the April 5th 2012 email chain where Dimon mentions “our exotic credit derivatives wind down”….See April 13th 2012 statements of Braunstein and Dimon “this is a senior management book run pursuant to the firm strategy”…To “help Hogan” ie “alleviate the pressure”, see the April 17th 2012 call where Drew calls Artajo actually and gives HER “guidance” to Artajo who said they had just come back within the “bid-offer”: “tweak that mark back” is the “guidance” of Drew as a response to Artajo….This could only mean that Artajo was instructed to set prices back OUT of the current “bid-offers” as per Drew latest directive, “if appropriate”…Of course….. Things were under full control really…. aside from Iksil’s unwanted alerts inside the firm…But under the direct watch of “Hogan”, the OCC and Dimon no doubt…Indeed, who can imagine that Drew told Artajo to stick outside of Bid-offer quotes the very day after she had met with the OCC, and the very day when “Hogan” needed help? But Drew allegedly will be “betrayed” about these CIO London Front Office prices, won’t she? And she will be “retired” with a golden parachute on May 14th 2012.

May 14th 2012 anyone on the trading floor of Jp Morgan New-York can shed a tear seeing the tender “bear hug” of Dimon to Drew as she is retired in haste and presented already as “the last victim to date of the London Whale”…. Was it the case really that she had been “betrayed” and therefore a “victim”? Late 2012 Drew’s emotional reversals on the distance spreadsheet and the March 20th P&L reports leave the US Senate commission wondering openly…. March 2013 statements of Drew under oath before the US Senate Commission: she was betrayed if one believes HER version of facts….The March 2013 analysis of US senate report does not concur with HER version at all…..This is hardly credible and puzzling in fact for the US Senate Commission when Drew testifies. September 2013 Carl Levin who chaired the commission stated: “« the whole issue of misinforming investors and the public is conspicuously absent from the SEC findings and settlement » “The size of the penalties is testimony to the great damage risky derivatives bets can do, and that's important. However, the whole issue of misinforming investors and the public is conspicuously absent from the SEC findings and settlement. Our PSI investigation showed that senior bank executives made a series of inaccurate statements that misinformed investors and the public as the London Whale disaster unfolded. Other civil and criminal proceedings apart from this settlement are continuing, so there is still time to determine any accountability on that matter.”

On the whole $6 Bln of a decoy “trading loss” were placed in the 10-Q reports by the very top of Jp Morgan against all realities prevailing outside of CIO but still within the walls of the banking group. All this seems to have been meant to “quietly hook” $60 billion of net tangible and lasting gains for the firm, about $25 billion being “done” in 2012 alone net of “collateral costs”. Yes this “London whale” scandal was just that: a plain “tempest in a teapot” knowingly so within the walls of Jp Morgan and organized at the very top. That “London whale” event had long been planned to be so by the bank: a trading scandal that would turn out to be just a “tempest in a teapot”, a “trading loss at CIO” in any event caused by some “traders” presumably so. Not only the bank made massive gains as projected earlier on but it cleaned its balance sheet for good after 10 years of patience. Mission accomplished isn’t it? Who cares about “traders” being lashed out?

It remains to see whether the total value generated in the build-up of Jp Morgan is really a gain for the economy as a whole and not actually a massive loss once adjusted for the inflation rate among other things in the future ….Small is beautiful.

179